Rodrigo Riquelme es ya una realidad. El modo en que Riquelme se hizo un llavero con Gorosabel antes de fusilar de puntera cual futbolista de fútbol sala fue una maravilla. Pero no hay que olvidar que

Así narró en directo Movistar el gol de Riquelme…¡los pelos de punta! | OneFootball

Without a “concerted” international effort, the amount of plastics seeping into the environment each year could double by 2040 – and reducing production at source is a key component of changing the direction of travel. That is according to a new ...

Report: Plastic pollution levels on course to double by 2040 without urgent action

Dan Harmon’s (Rick and Morty) new animated series set in a mythical version of Ancient Greece had three seasons ordered by Fox before it even aired.

TV & Movies Reviews

Electrolyser capacity for hydrogen production is likely to exceed 2GW by the end of the year, marking a year-on-year tripling, according to the International Energy Agency (IEA). The IEA revealed the figures this week through the new edition of its annual ...

IEA: Global green hydrogen pipeline tripled in 2023 but future growth at risk

Martyn Rickels and Simon Farr of specialist business advisory firm FRP Advisory were appointed as joint administrators to Aceleron Limited on 12 September 2023. Based in Bromsgrove, Aceleron has been trading since 2016 as a lithium battery developer that ...

Administrators in at lithium battery developer

A BROMSGROVE based battery technology company has gone into administration. Aceleron, an innovative lithium battery developer, appointed administrators, FRP Advisory, on September 12 after struggling to secure further investment. The company was founded in ...

Bromsgrove battery technology firm collapses into administration

Aceleron

aceleron

44,000 startups. 4,553 cities. “4 billion votes” It was such a blessing to get to know so many of your ideas and innovations from every corner of the world in 2021. The startup community amazed all of us with your hard work and persistence despite all of the challenges that we faced globally throughout last year. Not gonna lie, the competition got our hearts racing like Eurovision. We won’t let you wait for much longer, below are long-awaited results! ✨✨✨ To see who is a winner in each respective city, visit and pick a city. The name at the top of the list is the winner. (Compromised awards are under review.) startups.hackernoon.com The will announce the winners per region in their respective cities. Startups Winners Announcement Series 🔥 Central Europe Central Europe has always caught our attention with the talented pool of people in tech community. But were they able to get through 2021 despite the pandemic? Let’s cheers to the most outstanding amongst 4,500+ startups who have been working hard this year: Basel, Switzerland An overwhelming vote for ~ in their mission of helping machinery companies drive more service revenues from their installed base MachiQ Software Berlin, Germany We have with a whooping 538 votes! Let’s support Plan A with its mission to help companies to measure, monitor and reduce their environmental footprint. Plan A Düsseldorf, Germany It doesn’t get any easier with Kicking off in helping people organize online schedules more easily. EasyWeek! Geneva, Switzerland Manage code quality? We have to give props to for their startup! SonarSource Hamburg, Germany Taking over Hamburg is , an acoustic device to reduce noises in vehicle cabins. No more annoying sounds on the road 🚗 ReCalm Köln, Germany Small business? Hoping to manage your marketing campaign? is winning the vote! rankingCoach Lausanne, Switzerland is a Swiss-based company focused on cryptocurrency management solutions built for financial institutions. Metaco Leipzig, Germany empowers companies to operate and utilize drones for the most effective industrial data applications. FlyNex Mainz, Germany is growing sustainable digital mobility startups! Next Mobility Labs Munich, Germany wins for top-line, high-security international Identity Management! ORY Stuttgart, Germany Remote desktop application? Ask anywhere. AnyDesk Vienna, Austria , an IT company enabling privacy-preserving big data applications in the digital age. Mostly AI Zug, Switzerland Way to go 1762 votes for the first provably live pure Proof-of-Stake (PoS) CBC-Casper blockchain designed for real-world applications. CasperLabs! Nyon Calling for robust data solutions for impact-driven organizations! Havuta LLC Zürich is revolutionizing small business lending to give easy access to capital to those who need it most: SMEs ✨ Teylor Heidelberg, Germany is an innovative HR startup providing a solution for fast and efficient recruiting of employees with simple job profiles. Talk'n'Job Stuttgart, Germany Congratulations for winning this title! AnyDesk Moving on to our buddies in... 🔥 Eastern Europe Eastern Europe is home to many familiar big names in our start-up world. Trace history for Miro, Telegram, and Grammarly and you know where they are from. Who is the next generation for Eastern European startups? Budapest, Hungary is a platform for mobile app development, enabling tens of thousands of developers to automate integration, testing, and deployment. Bitrise I__ __ stanbul, Turkey Leading here is - a brand new software intelligence platform- helps organizations deliver faster, high-quality products by connecting their DevOps tools. Oobeya Kraków, Poland is a human cloud platform that connects global enterprises with European software houses, offering tech talent as a service. Talent Alpha Kyiv, Ukraine 281 votes for , an AI powered app where people can swap faces in the videos, GIFs and images in just a few seconds. Reface Sofia, Bulgaria 387 votes for __ __378, leading Blockchain and DLT solutions company that specializes in helping enterprises and blockchain startups design, build and scale innovative solutions using Blockchain & DLT! LimeChain Warsaw, Poland is a next-generation blockchain document management software (DMS). DoxyChain Yerevan, Armenia Whoever clicked to vote for was sure a vote, because it ended up winning as a full-service software provider delivering Artificial Intelligence & Machine Learning solutions for businesses. SmartClick smart Almaty Just like the name suggested, provides one pass to dozens of gyms and studios. 1FIT Debrecen got all the coders ranked figured out! CodersRank Kharkiv is complete solution digital assets management! Pics.io Minsk, Belarus with 243 votes is the first blockchain-powered sports media platform. Scorum Moscow, Russian Federation 1 e-grocery company in Russia with 26 votes and #1 on our HackerNoon board for Minsk. SberMarket Saint Petersburg, Russian Federation leads this area as a personal finance application that rewards responsible spending and savings. CoinKeeper Kazan, Russian Federation Smelling a sweet victory? 3d App designers for flower bouquets is in the house! Flowito Western Europe 🔥 Brexit? Pandemic? It seems like startups in Western Europe still keep their spirits regardless. Startups in Western Europe that stood out in votes this year go to: Antwerp is on a streak! As a bug bounty platform, it ensures online security through a crowd-sourced platform for computer experts. intigriti Belfast A record-breaking 10,003 votes? is an independent coffee shop. Come grab a coffee! Bobs & Berts Birmingham What’s happening in Birmingham? We have , an innovative UK-based advanced lithium battery developer aiming to accelerate the global shift to cleaner. Aceleron Bordeaux 30 is an open-source web framework for building modern apps and websites with Vue.js. NuxtJS Boulogne-billancourt is a software company that provides an in-store analytics platform to optimize operations. DiGEiZ Brighton In for some jam? offers musical education services. BIMM Bristol Another tasty victory for Western Europe is enabling the strongest digital encryption on the internet. KETS Quantum Security Brussels A fashionable win for , a social fashion network. You scan barcodes in fashion shops and see every influencer wearing those outfits! Share your wishlist with your friends and earn cashback. SCANDALOOK Cardiff is a technology company developing end-to-end digital connectivity for the point-of-care diagnostics market. Bond Digital Health London A whooping number of 1085 vote__ __ offers hardware based cyber-security solution for authentication and encryption in a wired local network. Crypto Quantique Edinburgh Hopping up on the crypto movement, just launched last year but they are already rocking it at making crypto accessible to everyone. Zumo Ghent 5379 votes for ! A Belgian InsurTech startup that helps insurance providers all over the world to better connect with their digital customers. WeGroup Glasgow specializes in single-cell analysis transforming the life sciences. Biomage Paris Show some love in the city of love for , a sports platform that is connected to a mobile sports coaching application. AirFit Aberystwyth enables artificial intelligence, data science & digital innovation solutions for the agriculture, biotech & life sciences industries. Agxio Norwich Congratulations for the amazing win of 1124 votes! WATR Charleroi Congratulations for your amazing win! Hirering Gibraltar Give a round of applause to Fortknoxter! Waltham Abbey is ruling the area with its amazing 72 votes! RPAIR Antwerpen Rocking up Antwerpen area is the rising startup ! Willow Egham What’s happening in Egham? Proudly announcing winning the Noonies this year! ATA.ONE Leicester Leicester is calling the name of as the winner for the region! DeadHappy Macclesfield Who else but ? Congratulations on dominating the award! AeroCloud Northern Europe 🔥 Want to listen to some music? Spotify. Need some games to kill time? Candy Crush. What about a quick call to someone you haven’t seen in a long time? Skype. These are all startups that originated from Northern Europe! Noonies, let’s take a look at the upcoming generation of startups from Northern Europe that hopefully will see an even bigger and brighter future. Gothenburg, Sweden Shipping was never made for small shippers, so took on the mission to built an end-to-end logistics platform to empower small businesses to spend less time and money on shipping and more time on growing their businesses. Sendify The Netherlands provides a virtual platform for many kinds of conferences and events. Let's Get Digital Helsinki, Finland A curated collection of public basketball courts in the Helsinki Metropolitan Area called . Everyone is welcome to submit to their local court. Koripallopaikat Lund, Sweden develops an antibiotic-free, safe and effective treatment of vaginal fungal infections. Gedea Biotech Luxembourg, Luxembourg 🚀 is a job-matching platform that connects developers and tech experts with CTOs and team leads from the best tech companies across Europe. 👾 nexten.io Malmö, Sweden Feeling studious, Noonies? __ __e is a tool for Google Classroom, that enables formative matrix assessment towards course learning objectives directly. StudyBe Oslo, Norway is the platform supports a unified API across all connected device categories, removing significant friction for app developers. Domos Labs Reykjavík, Iceland __ __s is a mobile-first technologically advanced neo-bank. Indo service #S__ __ tockholm, Sweden is the biggest blockchain start-up in Sweden. Double winner of the most prestigious nomination in Nordics 'NyTeknik 33-listan', winner of the Deloitte Fast 50, and now Noonies Award! ChromaWay AB T__ __ allinn, Estonia Passing 1000 votes is a fully digital and decentralized ecosystem for all actors in supply chain. OBORTECH Vilnius, Lithuania is an easy-to-use money app for cards and accounts. Via Genome, users can open a business or a merchant account or a personal one with dedicated IBANs, access transfers, exchange currency, process payments, and manage the funds. Genome Aalborg Digura A legaltech offering tenants fast, digital and reliable legal assistance Aarhus Partisia Market Design provides secure, simple, efficient solutions for auctions and exchanges realizing the potential of the market. Helsingborg is a hi-tech company developing cardiac occlusion devices and other vascular implants. Occlutech Tallinna is an AI-powered system for marketing automation in social networks. POSTOPLAN 🔥 Southern Europe Southern Europe’s economy went through a hard time this year due to a huge reduction in tourism. But our startups are of course always headstrong, from Spain to Italy and all the way through Greece. Who are the heroes rising amidst changes of Southern Europe? Skopje is a versatile software development company that solves the business challenges of SMBs, big brands, and brave hearts with awesome ideas. CodeChem Sliema Sounds like just a relaxing exhale. is a holistic health and wellbeing app built to allow users work on the development of their physical, mental and spiritual health Few Tajonar is a company of experienced business & industry entrepreneurs offering solutions to industries based on cutting-edge technologies. das-Nano Tarragona is a technology-based company focusing on the production of integrated microsystems. IMICROQ Terni is an ICT-based integrated solution that is able to modulate, respond, assist and learn to control most home activities. Mind Tirana is a video content and e-commerce platform for the Balkans built on top of an Albanian language specialized search engine. Gjirafa Torino develops and commercializes algorithms based on Artificial Intelligence to increase the autonomy of small satellite missions. AIKO T__ __ orres Buy online groceries like you would buy offline with personalized picking as fast as in 4 hours with Vedras 360hyper! Trento is a music-based platform built around streaming platforms that allow users to discover new emerging artists and make new friends. Brots 2021, what a YEAR. Congratulations to all of the startups that made it to the top winning list for Europe! Stay tuned for Startup awards in 2022 and let’s cheers to a new year of more innovative ideas, bold innovations, and robust creations from you, Europe 🌷 More Information: If you’re a winner, please answer ! If you’re a reader, participant, just a curious individual, consider answering one of our interviews . this interview here Read the FAQs about or . Startups here here To see all of the winners you may visit , pick any city and view the name that appears at the top. this site If you’re interested in partnering with us, we invite you to look through some of these or reach out to our business department. unique sponsorship options Perhaps creating your very own tickles your fancy or maybe you have a whole new idea for how we can partner! writing contest Thank you to all of our sponsors! We truly appreciate your contribution to the tech world and thus the future! The Startups Awards are sponsored by: , , , , , , , , , , Blockster Brex FreeCodeCamp LaunchDarkly Coil Udacity Udemy Vercel Asia Pacific Transgender Network Bybit Everscale Network This post was written by An Le and edited by Natasha Nel. If you have any questions or concerns or would like to report a discrepancy please email ellen@hackernoon.com.

Meet the Winners of Startups in Europe

When you create your own startup, one of the most important things to consider is where you will get your early funding from. When you begin your project, revenue is likely to be scarce or merely stable, rather than abundant and the success of your business may be centered around whether or not you can secure external investment to jumpstart your venture. With this in mind, we will be exploring seven European venture capital firms that are experts in their selected industries, who may be able to help you secure the funding that is necessary for your sustained success. SNÖ Ventures SNÖ Ventures are a Norweigan early-stage venture capital firm, that offers investments to technology-related startups and business owners. The team at SNÖ Ventures have extensive knowledge and experience in technology-related ventures and some have had their own businesses succeed. The firm has established offices in their native Norway, in Oslo and has also established a presence in Palo Alto, in the United States since their founding by Magne Uppman and Teodor Bjerrang during 2015. (READ MORE: 7 Thought Leaders You Should Be Following in 2019 ) The average size of a round of seed funding conducted by the firm is $1 Million. At this time, SNÖ Ventures have only made investments in their own country, with the industries of the companies invested in ranging from technology to marketing. Two of their notable investments were in in 2016 and in 2017. Both of these funding rounds netted the companies in question $1 Million in funding. Timely Inzpire.me Inzpire.me is a marketing startup from Norway that seeks to connect different brands with appropriate social media influencers who can collaborate with them. Timely is a time tracking app, geared towards freelancers, which allows them to automatically track the amount of time they take on a task, this allows them to accurately charge by the hour and to manage their time more effectively. 01 Ventures 01 Ventures are a British venture capital firm, based in the country’s capital, London. They specialize in providing early-stage investment for deep technology companies. The investment team for 01 Ventures are all specialists in their own area of technology, in their own right, which helps to guide their decisions when making investments. 01 Ventures predominantly invest into European startups, however, they have also invested in companies in the United States and China. (READ MORE: From Zero to Hero: Peter Thiel’s 3 Rules for Entrepreneurial Success ) 01 Ventures were founded in 2015 by a group of investors namely, Chris Haley, Correy Voo, Eeswaran Navaratnam and Ton van’t Noordende. The average size of one of the firm’s seed funding rounds stands at $1.875 Million. As it stands, the firm has currently invested in the seed rounds of two different companies. Firstly, they invested in on the 3rd of July 2017 and then they invested in on the 8th of August 2017. The level of funding received by Bibblio stood at $1.5 Million and for Teralytic the investment stood at $2.25 Million. Bibblio Teralytic Teralytic is a New York-based agriculture technology firm that seeks to help farmers lower their costs and increase their crop yield by making use of machine learning, advanced sensor technology and soil analysis. Bibblio is a London-based company that provides a set of tools to publishers and content creation companies, helping them to make their content more relevant and engaging to their target audiences. Inventure Inventure is a Finnish firm that offers early-stage funding to startups that can show they are either innovative or are working in high-tech sectors. The venture capital firm was founded in 2005 by Sami Lampinen and Timo Tirkkonen. Inventure market themselves as the “Nordic technology fund” with the main bulk of their investment focused on countries in Europe, with some of their operations taking place in China. The average size of one of their recent seed funding operations stands at around $900,000. Inventure has invested in a number of seed rounds in the past, with recent, notable examples of their investments coming in the form of their involvement with on the 21st of February 2018 and their leading investment into on the 4th of June 2018. Thirdpresence OME Health Thirdpresence is a company based in Helsinki that provides video advertising services, powered by artificial intelligence, which helps to allow companies to engage with their audiences on a larger scale. OME Health is a London-based company that offers preventative health coaching and analytics to its customers. Inventure sunk $1.3 Million into the venture. Superhero Capital Superhero Capital is a Finnish venture capital firm that specializes in investing in business at the early stages. They predominantly involve themselves in Business to Business software startups, with emphasis on the e-commerce, fintech, healthcare and industrial technology sectors. Superhero Capital was founded on the 6th of June 2015 by a group of entrepreneurs, including Jakob Storå, Juha Ruohonen, Jussi Harvela, and Moaffak Ahmed. The company primarily keeps its investments focused in their native Finland. In the last two years, their average investment into other company’s seed rounds totalled around $1.2 Million. (READ MORE: 7 Leadership Lessons from Scandinavia ) Some examples of their more recent, notable investments can be found in their leading investment into on the 27th of March 2018 and their leading investment into on the 1st of June, 2018. Valohai Codescoop Valohai is a Finnish startup that provides a system that can provide training and infrastructure for businesses that are looking to develop their own machine learning platforms. Codescoop is a Finnish business with Spanish origins that are seeking to solve the problem of cumbersome software stacks by providing open source software management and quality tracking services to their clients. Seedcamp Seedcamp is a European seed fund that is currently based in London, England. They are an established firm with experience of investing in early-stage technology firms that are looking to crack into the global market. Their primary sectors and involvements are in financial technology, artificial intelligence and software. Seedcamp was founded on the 1st of May, 2007 by both Saul Klein and Reshma Sohoni. The fund primarily centers their investments around companies in the European Union, however, they also have a notable presence in North America. The average seed funding raised for their more recent investments since November 2018 stands at $2.025 Million. In recent times, they have made a number of notable investments. Some examples of these are their $4 Million seed investment into Ezra on the 15th of November, 2018 and their $1.6 Million investment into on the 23rd of January, 2019. Gameye is an American health startup, based in New York that is seeking to revolutionize early cancer detection by utilizing artificial intelligence and high-tech medical imaging systems. Gameye is a Dutch gaming startup that is offering custom built server architecture for multiplayer, online video games. Ezra Mercia Fund Managers Mercia Fund Managers is a British venture capital firm, based in Birmingham in the West Midlands region of England. They primarily decide to invest in early-stage technology companies in the West Midlands, however, they also invest in other businesses around the UK. Startups hoping to receive investment must show Mercia that they can grow rapidly and that they are in a strong position with their concept regarding intellectual property, their technology and their products. The main sectors explored by Mercia are bio-science, software, engineering and digital entertainment. The firm was founded in 2010 by Mark Payton and Tim Hazell. On average, their investments into early-stage businesses total between £50,000 and £250,000 for seed funding rounds. In later rounds, they have guidelines which urge them to only invest up to £1 Million in each portfolio company. In the last year, they have made an investment into a number of companies with notable goals. For instance, on the 3rd of October 2018, they launched a leading investment into and then, on the 21st of November, 2018 they also invested in . Aceleron Technically Compaitable Aceleron is a British-based company that develop lithium-ion batteries, with an emphasis on trying to help the world move over to more renewable sources of energy by using sustainable batteries. Technically Compaitable is a company that offers software which can be used by recruiters for the IT sector to create tailored tests for their candidates to find the best person for the role they are recruiting for. Speedinvest Speedinvest is an Austria-based venture capital firm that is headquartered in Vienna. They invest in early-stage startups, with an emphasis on technology companies. Their main area of operation is centred around business in the European Union, however, they also have a strong global presence which includes operations in Silicon Valley. The main areas of investment carried out by Speedinvest involves fintech. deep tech, industrial tech, consumer tech and network effects. The firm was founded on the 1st of June, 2011 by a group of entrepreneurs which includes Daniel Keiper-Knorr, Erik Bovee, Michael Schuster, Oliver Holle, and Werner Zahnt. With regards to their level of funding for seed stage startups, they offer investment of up to 1 Million Euros. Speedinvest has involved themselves in a number of high-profile investments. On the 22nd of January, they launched their leading investment into fintech startup, during the company’s seed round. Furthermore, on the 15th of November, 2018 they launched a leading investment in the seed funding round for . TaxScouts Admix.in TaxScouts, as previously mentioned is a fintech startup based in the UK. The company mission is to assist accountants and UK taxpayers with preparing their tax online, in the wake of the UK government’s aim to make tax digital in the country. Admix.in is an industry first startup that allows developers of both augmented reality and virtual reality to specifically programme adverts into their content. They aim to make these adverts as seamless in the user experience as possible. Originally published at startupstash.com on February 14, 2019.

Top European VCs for Seed Funding | Startup Stash

Authors:
(1) Albert Gu, Machine Learning Department, Carnegie Mellon University and with equal contribution;
(2) Tri Dao, Department of Computer Science, Princeton University and with equal contribution. Table of Links Abstract and 1 Introduction 2 State Space Models 3 Selective State Space Models and 3.1 Motivation: Selection as a Means of Compression 3.2 Improving SSMs with Selection 3.3 Efficient Implementation of Selective SSMs 3.4 A Simplified SSM Architecture 3.5 Properties of Selection Mechanisms 3.6 Additional Model Details 4 Empirical Evaluation and 4.1 Synthetic Tasks 4.2 Language Modeling 4.3 DNA Modeling 4.4 Audio Modeling and Generation 4.5 Speed and Memory Benchmarks 4.6 Model Ablations 5 Discussion 6 Conclusion and References A Discussion: Selection Mechanism B Related Work C Mechanics of Selective SSMs D Hardware-aware Algorithm For Selective SSMs E Experimental Details and Additional Results 3.3 Efficient Implementation of Selective SSMs Hardware-friendly architectures such as convolutions (Krizhevsky, Sutskever, and Hinton 2012) and Transformers (Vaswani et al. 2017) enjoy widespread application. Here we aim to make selective SSMs efficient on modern hardware (GPU) as well. The selection mechanism is quite natural, and earlier works attempted to incorporate special cases of selection, such as letting ∆ vary over time in recurrent SSMs (Gu, Dao, et al. 2020). However, as previously mentioned a core limitation in the usage of SSMs is their computational efficiency, which was why S4 and all derivatives used LTI (non-selective) models, most commonly in the form of global convolutions. 3.3.1 Motivation of Prior Models We first revisit this motivation and overview our approach to overcome limitations of prior methods. • At a high level, recurrent models such as SSMs always balance a tradeoff between expressivity and speed: as discussed in Section 3.1, models with larger hidden state dimension should be more effective but slower. Thus we want to maximize hidden state dimension without paying speed and memory costs. • Note that the recurrent mode is more flexible than the convolution mode, since the latter (3) is derived from expanding the former (2) (Gu, Goel, and Ré 2022; Gu, Johnson, Goel, et al. 2021). However, this would require computing and materializing the latent state ℎ with shape (B, L, D, N), much larger (by a factor of N, the SSM state dimension) than the input x and output y of shape (B, L, D). Thus the more efficient convolution mode was introduced which could bypass the state computation and materializes a convolution kernel (3a) of only (B, L, D). • Prior LTI SSMs leverage the dual recurrent-convolutional forms to increase the effective state dimension by a factor of Nx (≈ 10 − 100), much larger than traditional RNNs, without efficiency penalties. 3.3.2 Overview of Selective Scan: Hardware-Aware State Expansion The selection mechanism is designed to overcome the limitations of LTI models; at the same time, we therefore need to revisit the computation problem of SSMs. We address this with three classical techniques: kernel fusion, parallel scan, and recomputation. We make two main observations: • The naive recurrent computation uses O(BLDN) FLOPs while the convolutional computation uses O(BLD log(L)) FLOPs, and the former has a lower constant factor. Thus for long sequences and not-too-large state dimension N, the recurrent mode can actually use fewer FLOPs. • The two challenges are the sequential nature of recurrence, and the large memory usage. To address the latter, just like the convolutional mode, we can attempt to not actually materialize the full state ℎ. The main idea is to leverage properties of modern accelerators (GPUs) to materialize the state ℎ only in more efficient levels of the memory hierarchy. In particular, most operations (except matrix multiplication) are bounded by memory bandwidth (Dao, Fu, Ermon, et al. 2022; Ivanov et al. 2021; Williams, Waterman, and Patterson 2009). This includes our scan operation, and we use kernel fusion to reduce the amount of memory IOs, leading to a significant speedup compared to a standard implementation. To avoid the sequential recurrence, we observe that despite not being linear it can still be parallelized with a work-efficient parallel scan algorithm (Blelloch 1990; Martin and Cundy 2018; Smith, Warrington, and Linderman 2023). Finally, we must also avoid saving the intermediate states, which are necessary for backpropagation. We carefully apply the classic technique of recomputation to reduce the memory requirements: the intermediate states are not stored but recomputed in the backward pass when the inputs are loaded from HBM to SRAM. As a result, the fused selective scan layer has the same memory requirements as an optimized transformer implementation with FlashAttention. Details of the fused kernel and recomputation are in Appendix D. The full Selective SSM layer and algorithm is illustrated in Figure 1. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Albert Gu, Machine Learning Department, Carnegie Mellon University and with equal contribution; (2) Tri Dao, Department of Computer Science, Princeton University and with equal contribution. Authors: Authors: (1) Albert Gu, Machine Learning Department, Carnegie Mellon University and with equal contribution; (2) Tri Dao, Department of Computer Science, Princeton University and with equal contribution. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 State Space Models 2 State Space Models 3 Selective State Space Models and 3.1 Motivation: Selection as a Means of Compression 3 Selective State Space Models and 3.1 Motivation: Selection as a Means of Compression 3.2 Improving SSMs with Selection 3.2 Improving SSMs with Selection 3.3 Efficient Implementation of Selective SSMs 3.3 Efficient Implementation of Selective SSMs 3.4 A Simplified SSM Architecture 3.4 A Simplified SSM Architecture 3.5 Properties of Selection Mechanisms 3.5 Properties of Selection Mechanisms 3.6 Additional Model Details 3.6 Additional Model Details 4 Empirical Evaluation and 4.1 Synthetic Tasks 4 Empirical Evaluation and 4.1 Synthetic Tasks 4.2 Language Modeling 4.2 Language Modeling 4.3 DNA Modeling 4.3 DNA Modeling 4.4 Audio Modeling and Generation 4.4 Audio Modeling and Generation 4.5 Speed and Memory Benchmarks 4.5 Speed and Memory Benchmarks 4.6 Model Ablations 4.6 Model Ablations 5 Discussion 5 Discussion 6 Conclusion and References 6 Conclusion and References A Discussion: Selection Mechanism A Discussion: Selection Mechanism B Related Work B Related Work C Mechanics of Selective SSMs C Mechanics of Selective SSMs D Hardware-aware Algorithm For Selective SSMs D Hardware-aware Algorithm For Selective SSMs E Experimental Details and Additional Results E Experimental Details and Additional Results 3.3 Efficient Implementation of Selective SSMs Hardware-friendly architectures such as convolutions (Krizhevsky, Sutskever, and Hinton 2012) and Transformers (Vaswani et al. 2017) enjoy widespread application. Here we aim to make selective SSMs efficient on modern hardware (GPU) as well. The selection mechanism is quite natural, and earlier works attempted to incorporate special cases of selection, such as letting ∆ vary over time in recurrent SSMs (Gu, Dao, et al. 2020). However, as previously mentioned a core limitation in the usage of SSMs is their computational efficiency, which was why S4 and all derivatives used LTI (non-selective) models, most commonly in the form of global convolutions. 3.3.1 Motivation of Prior Models 3.3.1 Motivation of Prior Models We first revisit this motivation and overview our approach to overcome limitations of prior methods. • At a high level, recurrent models such as SSMs always balance a tradeoff between expressivity and speed: as discussed in Section 3.1, models with larger hidden state dimension should be more effective but slower. Thus we want to maximize hidden state dimension without paying speed and memory costs. • Note that the recurrent mode is more flexible than the convolution mode, since the latter (3) is derived from expanding the former (2) (Gu, Goel, and Ré 2022; Gu, Johnson, Goel, et al. 2021). However, this would require computing and materializing the latent state ℎ with shape (B, L, D, N), much larger (by a factor of N, the SSM state dimension) than the input x and output y of shape (B, L, D). Thus the more efficient convolution mode was introduced which could bypass the state computation and materializes a convolution kernel (3a) of only (B, L, D). • Prior LTI SSMs leverage the dual recurrent-convolutional forms to increase the effective state dimension by a factor of Nx (≈ 10 − 100), much larger than traditional RNNs, without efficiency penalties. 3.3.2 Overview of Selective Scan: Hardware-Aware State Expansion 3.3.2 Overview of Selective Scan: Hardware-Aware State Expansion The selection mechanism is designed to overcome the limitations of LTI models; at the same time, we therefore need to revisit the computation problem of SSMs. We address this with three classical techniques: kernel fusion, parallel scan, and recomputation. We make two main observations: • The naive recurrent computation uses O(BLDN) FLOPs while the convolutional computation uses O(BLD log(L)) FLOPs, and the former has a lower constant factor. Thus for long sequences and not-too-large state dimension N, the recurrent mode can actually use fewer FLOPs. • The two challenges are the sequential nature of recurrence, and the large memory usage. To address the latter, just like the convolutional mode, we can attempt to not actually materialize the full state ℎ. The main idea is to leverage properties of modern accelerators (GPUs) to materialize the state ℎ only in more efficient levels of the memory hierarchy. In particular, most operations (except matrix multiplication) are bounded by memory bandwidth (Dao, Fu, Ermon, et al. 2022; Ivanov et al. 2021; Williams, Waterman, and Patterson 2009). This includes our scan operation, and we use kernel fusion to reduce the amount of memory IOs, leading to a significant speedup compared to a standard implementation. To avoid the sequential recurrence, we observe that despite not being linear it can still be parallelized with a work-efficient parallel scan algorithm (Blelloch 1990; Martin and Cundy 2018; Smith, Warrington, and Linderman 2023). Finally, we must also avoid saving the intermediate states, which are necessary for backpropagation. We carefully apply the classic technique of recomputation to reduce the memory requirements: the intermediate states are not stored but recomputed in the backward pass when the inputs are loaded from HBM to SRAM. As a result, the fused selective scan layer has the same memory requirements as an optimized transformer implementation with FlashAttention. Details of the fused kernel and recomputation are in Appendix D. The full Selective SSM layer and algorithm is illustrated in Figure 1. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Cutting-Edge Techniques That Speed Up AI Without Extra Costs

Əlavə Xərclər Olmadan Süni İntellektini Sürətləndirən Ən Müasir Texnikalar

Hochmoderne Techniken, die KI ohne Mehrkosten beschleunigen

Técnicas de vanguardia que aceleran la IA sin costes adicionales

تکنیک های پیشرفته ای که هوش مصنوعی را بدون هزینه اضافی سرعت می بخشد

Técnicas de vangarda que aceleran a IA sen custos adicionais

追加コストなしでAIを高速化する最先端技術

Teknika manafaingana ny AI tsy misy sarany fanampiny

Најсовремени техники кои ја забрзуваат вештачката интелигенција без дополнителни трошоци

අමතර වියදම් නොමැතිව AI වේගවත් කරන අති නවීන තාක්ෂණික ක්‍රම

Farsamooyinka Goynta-Edge ee Kor u qaadaya AI iyada oo aan kharash dheeraad ah lahayn

Teknikat më të fundit që përshpejtojnë AI pa kosto shtesë

Ekstra Maliyet Olmadan Yapay Zekayı Hızlandıran Son Teknoloji Teknikler

I have been writing code professionally for more than five years now. For the first four years, I never cared about the size of my pull requests (PRs). However, in the last year, I transitioned from submitting massive PRs with thousands of lines of changes to breaking them down into smaller, more manageable ones. The benefits of this shift have been immense, and in this blog, I’ll share those advantages. According to GitHub, a pull request is: A pull request is a proposal to merge a set of changes from one branch into another. In a pull request, collaborators can review and discuss the proposed set of changes before they integrate the changes into the main codebase. Essentially, a pull request is a way of collaborating; we should do everything possible to enhance this collaboration. One effective method to improve this collaboration is to keep PRs small. There is no universal definition to differentiate between a small and a large PR. Relying solely on the number of lines changed is insufficient, as auto-generated code and tests can inflate line counts. When I refer to small PRs in this article, I mean dividing a larger PR into multiple smaller, logically coherent PRs. Each smaller PR should be standalone, mergeable, and deployable. I do not advocate artificial splitting like splitting a PR into two, one containing all the code and another with just the tests, as this approach fails to yield any benefits I share below. Efficient Reviews: Ask a programmer to review 10 lines of code, he’ll find 10 issues. Ask him to do 500 lines and he’ll say it looks good. While humorous, this quote carries truth. Everyone is busy with their own work, and when you ask someone to review a PR, you are essentially requesting their time. Reviewing a PR requires the reviewer to switch context from their own work, and if the review takes a considerable amount of time, it can be challenging for them to return back to their work, potentially affecting their motivation and commitment to the review. Smaller PRs, which only take about 20–30 minutes to review, are much easier to tackle compared to those that can take 2–3 hours. Plus, larger PRs often lead to oversights because our attention spans can only handle so much, and jumping between lots of changes in a single PR can be confusing. From my experience, smaller PRs tend to get better feedback and lead to more meaningful design conversations. Faster Reviews: At this point, I’m thinking of adding my PR to my will in case it’s still in review after I’m gone. Longer PRs require a significant time commitment from reviewers, making them less likely to get attention — especially if they’re not tied to high-impact features. Small PRs on the other hand are reviewed quickly, as they demand less of the reviewer’s time and are less intrusive. This speed of reviews can be crucial for meeting project deadlines; I’ve seen projects getting delayed because senior reviewers couldn’t allocate time for massive PRs (though this can happen with small PRs, the risk is inherently higher with large ones). Lesser Rework: Reworking a big PR after design changes is like rearranging deck chairs on a ship you just finished building… and then sinking it. We have all experienced situations where someone realizes during PR review that a different design would have been more maintainable and future-proof, and we need to spend additional time reworking the PR based on the new design (not to say that they were completely on board with the design that was initially implemented :p). This is very natural because sometimes things become clearer once you see them written in code, and you start noticing aspects you might have missed during the design phase. With large PRs, this can be a significant issue because you have to rework a lot of elements, but with smaller PRs, it is easier to make changes. More importantly, there is a lower chance of rework because reviewers are more likely to identify issues early on and address them in the initial PRs, allowing subsequent PRs to be based on the new designs. Better Testing: Smaller PRs also benefit you as the author of the PR. They help in testing smaller changes incrementally rather than testing the whole project in one go. Testing smaller changes results in more exhaustive testing of each component of the system, thus leading to fewer production bugs. This applies to both automated tests and manual testing performed by you or dedicated QA engineers. Moreover, smaller PRs reduce the likelihood of missed test cases, as you can focus on a limited scope rather than the entire system. Writing tests? That sounds like Future Me’s problem. I have seen developers (including myself in the past) hesitate to write automation tests due to the perceived time investment without immediate, “visible” value to the feature/product. Smaller PRs reduce this friction by limiting the number of tests required and the time spent writing them. Easier Debugging: No matter how thorough your testing is, production bugs will happen! Being able to debug a bug in production is crucial since production bugs directly impact users, the business, or both. With large PRs, the surface area of change is also large, making it time-consuming and difficult to find the root cause of issues. On the other hand, smaller PRs contain less code and thus make debugging much faster. Debugging small changes is like finding a typo; debugging big changes is like proofreading an encyclopedia. Phased Feature Deployment: Last but not least, smaller PRs are also helpful to your product manager and users. By using small PRs, you can continuously push parts of the system to production, which helps in obtaining early feedback from users and allows for early course corrections if needed. Skipping early feedback is like cooking a five-course meal without tasting anything — you’re just hoping it’s not a disaster. The benefits of small PRs are numerous, and the points outlined above are among the most impactful I’ve experienced personally. If you have encountered other advantages of small PRs or challenges with larger ones that I haven’t covered, do comment with your insights. I hope this article motivates you to embrace smaller PRs. If you’re already on board, I hope it reinforces the value of this practice. Thanks for reading, until next time, keep coding and stay curious! I have been writing code professionally for more than five years now. For the first four years, I never cared about the size of my pull requests (PRs). However, in the last year, I transitioned from submitting massive PRs with thousands of lines of changes to breaking them down into smaller, more manageable ones. The benefits of this shift have been immense, and in this blog, I’ll share those advantages. According to GitHub , a pull request is: GitHub A pull request is a proposal to merge a set of changes from one branch into another. In a pull request, collaborators can review and discuss the proposed set of changes before they integrate the changes into the main codebase. A pull request is a proposal to merge a set of changes from one branch into another. In a pull request, collaborators can review and discuss the proposed set of changes before they integrate the changes into the main codebase. A pull request is a proposal to merge a set of changes from one branch into another. In a pull request, collaborators can review and discuss the proposed set of changes before they integrate the changes into the main codebase. Essentially, a pull request is a way of collaborating; we should do everything possible to enhance this collaboration. One effective method to improve this collaboration is to keep PRs small. There is no universal definition to differentiate between a small and a large PR. Relying solely on the number of lines changed is insufficient, as auto-generated code and tests can inflate line counts. When I refer to small PRs in this article, I mean dividing a larger PR into multiple smaller, logically coherent PRs. Each smaller PR should be standalone, mergeable, and deployable. I do not advocate artificial splitting like splitting a PR into two, one containing all the code and another with just the tests, as this approach fails to yield any benefits I share below. Efficient Reviews: Efficient Reviews: Ask a programmer to review 10 lines of code, he’ll find 10 issues. Ask him to do 500 lines and he’ll say it looks good. Ask a programmer to review 10 lines of code, he’ll find 10 issues. Ask him to do 500 lines and he’ll say it looks good. Ask a programmer to review 10 lines of code, he’ll find 10 issues. Ask him to do 500 lines and he’ll say it looks good. While humorous, this quote carries truth. Everyone is busy with their own work, and when you ask someone to review a PR, you are essentially requesting their time. Reviewing a PR requires the reviewer to switch context from their own work, and if the review takes a considerable amount of time, it can be challenging for them to return back to their work, potentially affecting their motivation and commitment to the review. Smaller PRs, which only take about 20–30 minutes to review, are much easier to tackle compared to those that can take 2–3 hours. Plus, larger PRs often lead to oversights because our attention spans can only handle so much, and jumping between lots of changes in a single PR can be confusing. From my experience, smaller PRs tend to get better feedback and lead to more meaningful design conversations. Faster Reviews: Faster Reviews: At this point, I’m thinking of adding my PR to my will in case it’s still in review after I’m gone. At this point, I’m thinking of adding my PR to my will in case it’s still in review after I’m gone. At this point, I’m thinking of adding my PR to my will in case it’s still in review after I’m gone. Longer PRs require a significant time commitment from reviewers, making them less likely to get attention — especially if they’re not tied to high-impact features. Small PRs on the other hand are reviewed quickly, as they demand less of the reviewer’s time and are less intrusive. This speed of reviews can be crucial for meeting project deadlines; I’ve seen projects getting delayed because senior reviewers couldn’t allocate time for massive PRs (though this can happen with small PRs, the risk is inherently higher with large ones). Lesser Rework: Lesser Rework: Reworking a big PR after design changes is like rearranging deck chairs on a ship you just finished building… and then sinking it. Reworking a big PR after design changes is like rearranging deck chairs on a ship you just finished building… and then sinking it. Reworking a big PR after design changes is like rearranging deck chairs on a ship you just finished building… and then sinking it. We have all experienced situations where someone realizes during PR review that a different design would have been more maintainable and future-proof, and we need to spend additional time reworking the PR based on the new design (not to say that they were completely on board with the design that was initially implemented :p). This is very natural because sometimes things become clearer once you see them written in code, and you start noticing aspects you might have missed during the design phase. With large PRs, this can be a significant issue because you have to rework a lot of elements, but with smaller PRs, it is easier to make changes. More importantly, there is a lower chance of rework because reviewers are more likely to identify issues early on and address them in the initial PRs, allowing subsequent PRs to be based on the new designs. Better Testing: Better Testing: Smaller PRs also benefit you as the author of the PR. They help in testing smaller changes incrementally rather than testing the whole project in one go. Testing smaller changes results in more exhaustive testing of each component of the system, thus leading to fewer production bugs. This applies to both automated tests and manual testing performed by you or dedicated QA engineers. Moreover, smaller PRs reduce the likelihood of missed test cases, as you can focus on a limited scope rather than the entire system. Writing tests? That sounds like Future Me’s problem. Writing tests? That sounds like Future Me’s problem. Writing tests? That sounds like Future Me’s problem. I have seen developers (including myself in the past) hesitate to write automation tests due to the perceived time investment without immediate, “visible” value to the feature/product. Smaller PRs reduce this friction by limiting the number of tests required and the time spent writing them. Easier Debugging: Easier Debugging: No matter how thorough your testing is, production bugs will happen! Being able to debug a bug in production is crucial since production bugs directly impact users, the business, or both. With large PRs, the surface area of change is also large, making it time-consuming and difficult to find the root cause of issues. On the other hand, smaller PRs contain less code and thus make debugging much faster. Debugging small changes is like finding a typo; debugging big changes is like proofreading an encyclopedia. Debugging small changes is like finding a typo; debugging big changes is like proofreading an encyclopedia. Debugging small changes is like finding a typo; debugging big changes is like proofreading an encyclopedia. Phased Feature Deployment: Phased Feature Deployment: Last but not least, smaller PRs are also helpful to your product manager and users. By using small PRs, you can continuously push parts of the system to production, which helps in obtaining early feedback from users and allows for early course corrections if needed. Skipping early feedback is like cooking a five-course meal without tasting anything — you’re just hoping it’s not a disaster. Skipping early feedback is like cooking a five-course meal without tasting anything — you’re just hoping it’s not a disaster. Skipping early feedback is like cooking a five-course meal without tasting anything — you’re just hoping it’s not a disaster. The benefits of small PRs are numerous, and the points outlined above are among the most impactful I’ve experienced personally. If you have encountered other advantages of small PRs or challenges with larger ones that I haven’t covered, do comment with your insights. I hope this article motivates you to embrace smaller PRs. If you’re already on board, I hope it reinforces the value of this practice. Thanks for reading, until next time, keep coding and stay curious!

The Big Power of Small Pull Requests: How They Improve Reviews and Speed Up Development

El gran poder de las pequeñas solicitudes de incorporación de cambios: cómo mejoran las revisiones y aceleran el desarrollo

小さなプルリクエストの大きな力: レビューを改善し、開発をスピードアップする方法

Nguya monene ya masengi ya kobenda ya mike: Ndenge nini kobongisaka botali mpe kosala ete bokoli ya mbangu

ພະລັງອັນໃຫຍ່ຫຼວງຂອງການຮ້ອງຂໍດຶງຂະຫນາດນ້ອຍ: ເຮັດແນວໃດພວກເຂົາປັບປຸງການທົບທວນຄືນແລະເລັ່ງການພັດທະນາ

Nelielu pieprasījumu lielais spēks: kā tie uzlabo pārskatus un paātrina attīstību

Ny Hery lehibe amin'ny fangatahan'ny fisintonana kely: ny fomba hanatsarana ny hevitra sy ny fanafainganana ny fampandrosoana

सानो पुल अनुरोधहरूको ठूलो शक्ति: कसरी तिनीहरूले समीक्षाहरू सुधार गर्छन् र विकासलाई गति दिन्छ

Wielka moc małych żądań ściągnięcia: jak usprawniają przeglądy i przyspieszają rozwój

Fuqia e madhe e kërkesave për tërheqje të vogla: Si përmirësojnë rishikimet dhe përshpejtojnë zhvillimin

The Big Power of Small Pull Requests: Hur de förbättrar recensioner och påskyndar utvecklingen

พลังอันยิ่งใหญ่ของคำขอ Pull Request ขนาดเล็ก: วิธีปรับปรุงบทวิจารณ์และเร่งการพัฒนา

小型 Pull 请求的巨大威力：它们如何改善评审并加快开发速度

APAC's Bold Stride into the Future of Cryptocurrency As the digital asset ecosystem matures globally, the Asia Pacific (APAC) region emerges as a significant frontier for cryptocurrency and blockchain development. The region's evolving regulatory frameworks are beckoning a myriad of players eager to secure their place in this flourishing sector. Hong Kong stands out, leveraging its as a global financial nexus and gateway to APAC's diverse markets to attract firms that are eager to deliver cutting-edge digital asset services. The city's regulatory environment is earning accolades for its crypto-friendly stance, drawing in businesses ready to tap into Asia's burgeoning demand for avant-garde financial solutions. stature A wave of momentum in adopting local payment systems, including Hong Kong’s Faster Payment Network, underscores the region's nimble adaptation to new financial technologies. This not only caters to the growing appetite for Web3 services but also signifies a wider acceptance of the transformative power of decentralized finance. Conversations with industry veterans reveal a consensus: demystifying Web3 platforms is key to broadening the embrace of digital currencies in APAC. Streamlining the intricacies of crypto transactions and cryptographic protocols is drastically lowering the barriers to entry, making the technology accessible to a broader audience. A staunch dedication to regulatory adherence and international standards is at the forefront of these ventures. By meeting stringent compliance requirements, such as obtaining necessary licenses and aligning with financial oversight regulations, these companies are not just adhering to the law—they're championing best practices for the sector at large. Transak's Strategic Leap into Hong Kong: Fueling APAC's Web3 Revolution In an era where digital assets and decentralized finance are reshaping global markets, Transak announces its bold move into Hong Kong, marking a significant milestone in the advancement of the APAC region's nascent digital asset ecosystem. A New Dawn in Digital Finance Transak, renowned for its innovative Web3 payments and onboarding solutions, has chosen Hong Kong as its new home base, aiming to harness the city's unique position as a global financial nexus and doorway to the Asia Pacific's burgeoning market. This decision follows a triumphant Series-A funding round in mid-2023, spotlighted by the backing of heavyweight investors like Animoca Brands. Hong Kong: The Gateway to APAC's Web3 Future Hong Kong's strategic allure lies in its progressive regulatory framework for cryptocurrencies and digital assets, making it an ideal locale for Transak's ambitions. The firm plans to knit itself into the fabric of Hong Kong's financial ecosystem, enhancing local engagement through the adoption of popular payment methods and integration with the city's Faster Payment Network. This move is more than geographical; it's a strategic alignment with Transak's mission to demystify Web3 accessibility. Bridging Communities: Transak's Strategic Alliance with Web3 Harbour: Joining forces with Web3 Harbour, one of Hong Kong’s leading Web3 communities, Transak is not just expanding geographically but also embedding itself deeply into the local Web3 culture. This partnership underscores a mutual commitment to accelerate the adoption of blockchain technologies and the decentralized internet, facilitating cross-sector industry collaboration. Gary Liu, Chairman of Web3 Harbour, expressed enthusiasm for this union, marking it as a positive stride towards enriching the digital asset ecosystem in the APAC region. Strategic Partnerships and Regulatory Excellence: The Transak Way Transak is no stranger to the rigors of regulatory compliance, operating in over 150 countries with a commitment to adherence across an extensive array of regulatory landscapes. With an eye on the intricate dance of local compliance, Transak is recruiting a team of local directors and experts in Hong Kong, ensuring smooth operations and active engagement with policymakers. The expansion is buoyed by strategic collaborations, notably with Animoca Brands, a key player in Transak's Series-A funding round. Yat Siu, co-founder of Animoca Brands, heralds Transak's Hong Kong venture as a pivotal step towards harnessing the immense potential of the APAC market, aiming to onboard a billion users into the Web3 fold. Closing Thoughts Transak's foray into Hong Kong is not just a business expansion; it's a visionary step towards realizing the full potential of the digital asset economy in the APAC region. By bridging the gap between traditional finance and crypto assets, Transak is poised to catalyze the next wave of Web3 enthusiasts, making the decentralized internet more accessible and engaging for everyone. Don’t forget to like and share the story! Vested Interest Disclosure: This author is an independent contributor publishing via our brand-as-author program. Be it through direct compensation, media partnerships, or networking, the author has a vested interest in the company/ies mentioned in this story. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR

Transak's Bold Move: Accelerating Hong Kong's Journey to Web3 Dominance

ট্রান্সাকের সাহসী পদক্ষেপ: ওয়েব3 আধিপত্যে হংকংয়ের যাত্রাকে ত্বরান্বিত করা

Transaks mutiger Schritt: Beschleunigung von Hongkongs Weg zur Web3-Dominanz

La audaz medida de Transak: acelerar el viaje de Hong Kong hacia el dominio de Web3

Le geste audacieux de Transak : accélérer le parcours de Hong Kong vers la domination du Web3

ट्रांसक का साहसिक कदम: हांगकांग की वेब3 प्रभुत्व की यात्रा को तेज करना

トランザックの大胆な行動: 香港の Web3 優位への旅を加速する

Transak의 대담한 움직임: Web3 지배력을 향한 홍콩의 여정 가속화

A jogada ousada da Transak: acelerando a jornada de Hong Kong para o domínio da Web3

Смелый шаг Transak: ускорение пути Гонконга к доминированию Web3

Transak'ın Cesur Hareketi: Hong Kong'un Web3 Hakimiyeti Yolculuğunu Hızlandırma

Bước đi táo bạo của Transak: Đẩy nhanh hành trình thống trị Web3 của Hồng Kông

Transak 的大胆举措：加速香港迈向 Web3 主导地位

Opside is a decentralized ZK-RaaS ( as a Service) platform and a PoW network that supports ZKP (Zero-Knowledge Proof) mining. Opside uses a hybrid PoS & PoW consensus and provides a one-click feature for Web3 developers to create zkEVM application chains. At the same time, the massive computational tasks brought by many ZK-Rollups have generated a huge demand for computational power, providing a meaningful mining scenario for miners around the world. ZK-Rollup Opside's mission is to make ZK-Rollup more user-friendly and popular, thereby enabling the widespread implementation of application chains based on zkEVM. Opside provides the following services and innovations: ZK-Rollup as a Service: For developers, they can deploy their own ZK-Rollup (zkEVM) on multiple chains with one click. Opside ZK-PoW Cloud: For miners, Opside is also a multi-chain PoW protocol, supporting ZK mining, and generating zero-knowledge proofs for ZK-Rollups on multiple public chains. Advantages of Opside According to Vitalik's latest article, we will transition from a multi-chain era to a multi-Rollup era in the future. The ZK-Rollup as a Service provided by Opside will make it possible for ZK-Rollup to be widely used. Developers do not need any ZK knowledge, they can create their own ZK-Rollup by selecting through the front-end interface. The created ZK-Rollup can become the Layer 2 of any public chain (such as Ethereum, BNB Chain, or Polygon PoS). The computational power involved will also be provided by the Opside ZK-PoW multi-chain protocol. This greatly reduces the development and operational costs of the project. In addition, Opside will launch specific contracts with 0 gas fees in early July. When users use certain dApps on some custom ZK-Rollup, they will not have to pay any gas fees. These 0 gas fee contracts aim to solve the current high entry barriers of Web3 dApps and greatly improve the user experience. Value of Opside ZK Ecosystem Since the on May 24, 2023, the Opside network has received significant market attention. Within just a month, there are over 120,000 unique addresses across the network, with a transaction volume reaching 5 million, more than 12,000 PoS Validators, and 60+ miners. The cost value injected into Opside by global community members has already reached 25 million dollars. Opside Pre-alpha Testnet was launched Moreover, it is worth noting that the ZK application ecosystem of Opside is also continuously expanding. So far, six projects have created sovereign ZK-Rollups and deployed related applications, and more than ten other projects have been deployed on the Opside Public zkEVM. In addition, more than 25 projects are in the process of docking and scheduling. The ecosystem projects involve multiple popular tracks such as , GameFi, SocialFi, Content, NFT, etc. DeFi Here are some ecosystem project introductions: Sideswap Intro: The leading dex and liquidity marketplace on Opside powered by ZK-RaaS. Category: Dex, Defi infra Chain: Opside Chain, Opside public zkEVM V2 Galxe campaign: (Potential airdrop) https://galxe.com/Opside/campaign/GCeDcUSZZ5 Cryptoblades Intro: CryptoBlades is a revolutionary web-based NFT roleplaying game. Category: Game Chain: CB zkEVM (custom zkEVM) Galxe campaign: https://galxe.com/Opside/campaign/GCDK8UWsgJ Era7: Intro: Era7 Game of Truth is a metaverse-style TCG. Category: Game Chain: ERA7 (Custom zkEVM) Galxe campaign: https://galxe.com/Opside/campaign/GCoQcUS3Ms BlockNG: Intro: BlockNG aims to connect every participant in the classical blockchain & build a strong ecosystem. Category: Dex, Defi infra, NFT Chain: Opside public zkEVM V2, Law Chain (Custom zkEVM) Galxe campaign: https://galxe.com/Opside/campaign/GCcUXUX8S9 XThrill Intro: XThrill is a causal game blending high-speed racing with intense shootouts, harnessing #ZK-Rollup technology. Category: Game Chain: XThrill (Custom zkEVM) Galxe campaign: (Potential airdrop) https://galxe.com/Opside/campaign/GCwBqUSYsz iZUMi Finance Intro: iZUMi Finance is a multi-chain DeFi protocol providing one-stop Liquidity as a Service (LaaS). Category: Defi Chain: Opside public zkEVM V2 SoQuest Intro: SoQuest is an all-in-one Web3 Quest platform that helps projects build Web3 communities and aggregate various project activities and information. Category: Tool Chain：SoQuest zkEVM（Custom zkEVM） ReadON Intro: ReadON is a decentralized content distribution platform. Category: Content Chain: Custom zkEVM (Live soon!) Elfin Kingdom Intro: Elfin Kingdom is an MMORPG interplanetary kingdom in the Metaverse. Category: Game Chain: Custom zkEVM (Live soon!) Ivy market Intro: All-encompassing NFT marketplace that enables you to unlock additional benefits. Category: NFT Marketplace Chain: Opside public zkEVM V2 Jackbot Intro: Win-2-earn classic games with web3 tech. Category: Game Chain: Jackbot(Custom zkEVM) Benefits of Deploying on Opside Opside is EVM-compatible, meaning that any project developed with the Solidity language can be migrated and deployed at nearly no cost. There's also no learning curve for new projects. Project owners can enjoy the security offered by ZK-Rollup, low gas fees, minimal operational costs, and greatly improved user experience. Here are eight major advantages of Opside: ZK-RaaS: Opside provides ZK-Rollup as a service with enhanced security, trustlessness, and faster withdrawals. Support Ethereum/Opside Chain/BNB Chain/Polygon and more base chains. Managed Hardware: Required ZK-Rollup hardware resources are offered in a decentralized, permissionless way, reducing costs for developers. Zero Gas Fee: Opside's custom Rollup Chains offer unique benefits when it comes to the economic model for fees. Developers are free to adjust the gas fee, even to 0. Thus, users do not need to pay any fees. Fully On-Chain: The platform facilitates fully on-chain implementation of games, financial derivatives, and social networks due to low gas fees and high throughput. EVM Compatibility: Opside supports various zkEVMs, offering a free choice of zkEVM SDK for developers and eliminating migration costs. Sovereignty: Developers retain sovereignty over ZK-Rollup, including the ability to customize the economic model and implement zero gas fees. Cross-Rollup Communication: A unified system contract facilitates native cross-rollup communication in different ZK-Rollups with the same basechain. Decentralization: Utilizing ETH 2.0's PoS consensus, Opside enables more extensive validation and miner participation, enhancing security. The Potential of Opside More and more developers and project owners have demands for sovereign ZK-Rollup and 0 gas fee. Project owners can generate sovereign ZK-Rollups via Opside's official rollup launchbase in 10 minutes, along with setting up basic matching settings such as browsers and bridges. Sovereign ZK-Rollup means higher security and stability. The 0 gas fee reduces user entry barriers, allowing many Web2 traffic to enter Web3 without obstacles. After transitioning from ETH to PoS, many miners have no place to go. Opside's multi-chain ZKP PoW protocol can connect any rollup, providing them with decentralized computing power to generate ZK proofs. Opside maintains good relationships with global miners like Ingoyama and others, with joint promotions already underway. At present, Opside is still in the early stage. It has been more than a month since the testnet was launched, and it already has over 12k validators, over 60 large miners, and more than 120k independent user addresses. A large-scale increase in various data is expected in the future. Conclusion & Outlook As a decentralized ZK-RaaS platform and a PoW network that supports ZKP mining, Opside has accelerated the widespread implementation of application chains based on zkEVM. Opside not only provides a one-click ZK-Rollup application chain generation feature, reducing the technical burden for developers, but it also uses its ZK-PoW Cloud to meet the computational demands of ZKP tasks generated by the increase in ZK-Rollups, creating a new mining scenario for global miners. Moreover, Opside's 0 gas fee contract greatly enhances user experience and reduces entry barriers, which is of epoch-making significance in solving the current high entry threshold of Web3 dApps. Since the launch of the Opside Pre-alpha Testnet on May 24, 2023, it has gained widespread attention and positive feedback globally. With the continuous advancement and development of Rollup, there is reason to expect that ZK-Rollup technology will become more prevalent in many application scenarios. The emergence and development of Opside have made the large-scale implementation of ZK-Rollup possible. Also, it provides new mining opportunities for miners. Opside maintains a good cooperative relationship with large miners globally, laying a solid foundation for Opside's future development. Although Opside is currently in the early stage, its performance on the testnet has already demonstrated its huge potential. In the future, as more developers and project owners join, it is expected that Opside's ecosystem will further expand, and the transaction volume and user base will continue to grow. Join the Journey! Opside invites more project owners to join our ZK ecosystem. Generate sovereign ZK-Rollup application chains with one click through launchbase. Whether it's a new start-up project or an already stable operating project, they can get the necessary support and convenience from Opside. Opside not only helps project owners quickly deploy and operate but also provides continuous technical support and an ecosystem development environment to maximize the value of the project. At the same time, Opside also sincerely invites all users to explore and discover projects they are interested in within our ecosystem. The Opside network ecosystem brings together applications from various popular tracks for community users to experience. We hope Opside can bring users a new interaction experience and allow users to participate and experience the charm of ZK more deeply.

Opside Ecosystem: Growing Together & Accelerating the Multi-Rollup Era

Ecosistema Opside: creciendo juntos y acelerando la era de los paquetes múltiples

Opside Ecosystem : grandir ensemble et accélérer l'ère du multi-rollup

ऑप्ससाइड इकोसिस्टम: एक साथ बढ़ना और मल्टी-रोलअप युग में तेजी लाना

Opside エコシステム: 共に成長し、マルチロールアップ時代を加速する

Ecossistema Opside: Crescendo Juntos e Acelerando a Era Multi-Rollup

Hệ sinh thái Opside: Cùng nhau phát triển & Tăng tốc Kỷ nguyên Multi-Rollup

Opside 生态系统：共同成长，加速 Multi Rollup 时代

The Graph Network has upgraded its support for the Solana blockchain to improve the development process for decentralized applications (dApps) on Solana. These enhancements provide Solana developers with better data tools and increased access to indexed Solana data through pre-built solutions from teams like Messari and Top Ledger. The Graph Network has introduced upgrades focused on Substreams-powered subgraphs (SpS). New code generation (codegen) tools developed by StreamingFast, part of The Graph ecosystem, are now available for both remote and local use. These tools help developers set up end-to-end Substreams projects more efficiently, allowing faster synchronization for those familiar with Substreams. Increasing Accessibility A notable aspect of the latest upgrades is the increased accessibility for developers who may not have prior experience with Substreams or Rust. The Graph has implemented tooling that allows developers to interact with Solana data without needing to engage directly with these specific technologies. This approach broadens the potential user base, enabling developers to leverage Solana’s capabilities using tools that align with their existing skills and project requirements. By lowering these barriers, The Graph fosters a more inclusive development environment, encouraging a diverse range of projects on the Solana network. Expanded Toolset and Pre-Built Solutions In addition to the improved Substreams tooling, The Graph has integrated Solana Substreams modules developed by Top Ledger and StreamingFast. These modules, along with multiple Substreams-powered subgraphs created by Messari, provide robust solutions for data indexing and retrieval. Developers can also utilize ready-made subgraphs, such as those for UNCX and Orca Whirlpool, which significantly reduce the time and effort required to implement complex functionalities in dApps. This expanded toolset not only accelerates development but also ensures that applications can access reliable and up-to-date data efficiently. Nick Hansen, Head of Growth at The Graph Foundation, commented on the upgrades: “The increase in developer and user activity on Solana has raised the demand for open, decentralized data. The Graph’s latest tooling upgrade and enhanced support will help the Solana community utilize the decentralized data layer of Web3 more effectively.” This feedback highlights the alignment between The Graph’s enhancements and the evolving needs of the Solana ecosystem, emphasizing the importance of robust data infrastructure in supporting rapid growth. Addressing Solana’s Needs Solana has gained traction in the blockchain space due to its high transaction speeds and low fees, making it suitable for a wide range of dApps and decentralized finance (DeFi) projects. As the Solana ecosystem continues to expand rapidly, the need for efficient data indexing and querying solutions has surged. The Graph’s upgrades aim to meet this demand by reducing costs, improving data sources, and enhancing the indexing stack. Additionally, The Graph now offers a historical cache of Solana blocks that is 25% the size of a full block, optimizing storage and retrieval processes. These improvements are crucial for maintaining performance and scalability as the volume of data and number of applications on Solana grow. The Graph’s upgrades are part of its broader mission to support a multi-chain future and enhance interoperability within the Web3 ecosystem. By refining its tooling for Solana, The Graph not only improves the developer experience on one of the leading blockchains but also contributes to the overall scalability and functionality of decentralized applications across different networks. This multi-chain support is essential in an increasingly interconnected blockchain landscape, where applications often need to interact with multiple platforms to deliver comprehensive solutions. Transition to Decentralization Earlier this year, The Graph Network transitioned from a free hosted service to a fully decentralized model. This shift ensures data autonomy and decentralization for developers and the broader Web3 ecosystem. With over 9,000 subgraphs in use, The Graph provides a scalable data infrastructure for developers on various blockchains, including Ethereum, Arbitrum, Optimism, Base, Polygon, Celo, Fantom, Gnosis, and Avalanche. This move reinforces The Graph’s commitment to decentralization, offering a more resilient and secure data indexing service that aligns with the foundational principles of Web3. By providing enhanced tooling and easier access to indexed data, The Graph lowers the technical barriers for developers, potentially increasing the number and diversity of dApps built on Solana. This can lead to a more vibrant and competitive ecosystem, attracting more users and investors. Moreover, the ability to efficiently index and query large volumes of data is critical for the growth of DeFi applications, which rely heavily on real-time data for functionalities such as lending, borrowing, and trading. The improved latency and scalability offered by FTSOv2 ensure that these applications can operate smoothly even as user activity and transaction volumes increase. The transition to a fully decentralized model also enhances the security and reliability of data services, which is paramount for maintaining user trust and safeguarding against potential data breaches or manipulations. As blockchain technology continues to evolve, robust data indexing solutions like those provided by The Graph will be integral to supporting innovative use cases and driving the next wave of decentralized applications. The Graph’s latest upgrades for Solana developers represent a significant advancement in the tools and resources available for building decentralized applications. By improving latency, scalability, and cost-efficiency, The Graph addresses the critical needs of the growing Solana ecosystem and supports the development of Web3 solutions. These enhancements not only benefit Solana developers but also contribute to a more interconnected and efficient decentralized web, fostering innovation and broader adoption of blockchain technology. Don’t forget to like and share the story! Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR The Graph Network has upgraded its support for the Solana blockchain to improve the development process for decentralized applications (dApps) on Solana. These enhancements provide Solana developers with better data tools and increased access to indexed Solana data through pre-built solutions from teams like Messari and Top Ledger. Graph Network The Graph Network has introduced upgrades focused on Substreams-powered subgraphs (SpS). New code generation (codegen) tools developed by StreamingFast, part of The Graph ecosystem, are now available for both remote and local use. These tools help developers set up end-to-end Substreams projects more efficiently, allowing faster synchronization for those familiar with Substreams. Increasing Accessibility A notable aspect of the latest upgrades is the increased accessibility for developers who may not have prior experience with Substreams or Rust. The Graph has implemented tooling that allows developers to interact with Solana data without needing to engage directly with these specific technologies. This approach broadens the potential user base, enabling developers to leverage Solana’s capabilities using tools that align with their existing skills and project requirements. By lowering these barriers, The Graph fosters a more inclusive development environment, encouraging a diverse range of projects on the Solana network. Expanded Toolset and Pre-Built Solutions In addition to the improved Substreams tooling, The Graph has integrated Solana Substreams modules developed by Top Ledger and StreamingFast. These modules, along with multiple Substreams-powered subgraphs created by Messari, provide robust solutions for data indexing and retrieval. Developers can also utilize ready-made subgraphs, such as those for UNCX and Orca Whirlpool, which significantly reduce the time and effort required to implement complex functionalities in dApps. This expanded toolset not only accelerates development but also ensures that applications can access reliable and up-to-date data efficiently. Nick Hansen, Head of Growth at The Graph Foundation, commented on the upgrades: “The increase in developer and user activity on Solana has raised the demand for open, decentralized data. The Graph’s latest tooling upgrade and enhanced support will help the Solana community utilize the decentralized data layer of Web3 more effectively.” This feedback highlights the alignment between The Graph’s enhancements and the evolving needs of the Solana ecosystem, emphasizing the importance of robust data infrastructure in supporting rapid growth. Addressing Solana’s Needs Solana has gained traction in the blockchain space due to its high transaction speeds and low fees, making it suitable for a wide range of dApps and decentralized finance (DeFi) projects. As the Solana ecosystem continues to expand rapidly, the need for efficient data indexing and querying solutions has surged. The Graph’s upgrades aim to meet this demand by reducing costs, improving data sources, and enhancing the indexing stack. Additionally, The Graph now offers a historical cache of Solana blocks that is 25% the size of a full block, optimizing storage and retrieval processes. These improvements are crucial for maintaining performance and scalability as the volume of data and number of applications on Solana grow. The Graph’s upgrades are part of its broader mission to support a multi-chain future and enhance interoperability within the Web3 ecosystem. By refining its tooling for Solana, The Graph not only improves the developer experience on one of the leading blockchains but also contributes to the overall scalability and functionality of decentralized applications across different networks. This multi-chain support is essential in an increasingly interconnected blockchain landscape, where applications often need to interact with multiple platforms to deliver comprehensive solutions. Transition to Decentralization Earlier this year, The Graph Network transitioned from a free hosted service to a fully decentralized model. This shift ensures data autonomy and decentralization for developers and the broader Web3 ecosystem. With over 9,000 subgraphs in use, The Graph provides a scalable data infrastructure for developers on various blockchains, including Ethereum, Arbitrum, Optimism, Base, Polygon, Celo, Fantom, Gnosis, and Avalanche. This move reinforces The Graph’s commitment to decentralization, offering a more resilient and secure data indexing service that aligns with the foundational principles of Web3. By providing enhanced tooling and easier access to indexed data, The Graph lowers the technical barriers for developers, potentially increasing the number and diversity of dApps built on Solana. This can lead to a more vibrant and competitive ecosystem, attracting more users and investors. Moreover, the ability to efficiently index and query large volumes of data is critical for the growth of DeFi applications, which rely heavily on real-time data for functionalities such as lending, borrowing, and trading. The improved latency and scalability offered by FTSOv2 ensure that these applications can operate smoothly even as user activity and transaction volumes increase. The transition to a fully decentralized model also enhances the security and reliability of data services, which is paramount for maintaining user trust and safeguarding against potential data breaches or manipulations. As blockchain technology continues to evolve, robust data indexing solutions like those provided by The Graph will be integral to supporting innovative use cases and driving the next wave of decentralized applications. The Graph’s latest upgrades for Solana developers represent a significant advancement in the tools and resources available for building decentralized applications. By improving latency, scalability, and cost-efficiency, The Graph addresses the critical needs of the growing Solana ecosystem and supports the development of Web3 solutions. These enhancements not only benefit Solana developers but also contribute to a more interconnected and efficient decentralized web, fostering innovation and broader adoption of blockchain technology. Don’t forget to like and share the story! Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program . HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR Vested Interest Disclosure: business blogging program business blogging program

The Graph Upgrades Tooling for Solana Developers to Accelerate DApp Deployment

DApp tətbiqini sürətləndirmək üçün Solana Tərtibatçıları üçün Qrafik Təkmilləşdirmə Alətləri

Графиката надгражда инструменти за разработчици на Solana за ускоряване на внедряването на DApp

The Graph actualiza las herramientas para que los desarrolladores de Solana aceleren la implementación de DApp

The Graph met à niveau les outils pour les développeurs Solana afin d'accélérer le déploiement des DApp

A Graph Upgrade Tooling a Solana fejlesztők számára a DApp telepítésének felgyorsítása érdekében

The Graph、Solana開発者向けツールをアップグレードし、DAppの展開を加速

The Graph, Solana 개발자를 위한 도구 업그레이드로 DApp 배포 가속화

Graph ulepsza narzędzia dla programistów Solana, aby przyspieszyć wdrażanie DApp

DApp යෙදවීම වේගවත් කිරීම සඳහා Solana සංවර්ධකයින් සඳහා ප්‍රස්ථාරය වැඩිදියුණු කිරීමේ මෙවලම්

Grafen uppgraderar verktyg för Solana-utvecklare för att påskynda DApp-distribution

Switirhisiwa swa ku Ndlandlamuxa Girafamu swa Vatumbuluxi va Solana ku Hatlisisa ku Tirhisiwa ka DApp

IsiXhobo soPhuculo lweGrafu kubaPhuhlisi beSolana ukukhawulezisa ukusasazwa kwe-DApp

In this article, I want to tell you about a useful plugin ADB Idea. I use it every day, but I noticed that others use it in their work quite rarely. Working With ADB Idea The plugin is off by default. After adding it, the toolbars will appear: All this plugin does is replace the standard ADB commands. For example, the is analogous to the terminal command. The advantage of ADB Idea is that the most frequently used commands are placed in the toolbar and you don't have to remember them all the time. ADB Kill App ADB shell am kill Start the app. Click on the command, and verify that the application has closed and disappeared from the recently used stack. ADB Kill App You can optimize even more. Since the commands are present in the panel, you can assign keyboard shortcuts to them, just like any other, which will also speed up development. Let's go into the Keymap settings: In the search box, type ADB and you will see the necessary commands. Find the previously described and assign a key combination to it, such as . Click OK. Close the menu, launch the app, and press . We will make sure that the application is closed. ADB Kill App Option+K Option+K Creating an ADB quick list There are quite a lot of useful adb commands. So, we run the risk of cramming all the key combinations. The way out of this is to create a quick list. Let's go to Preferences > Appearance & Behavior > Quick Lists: Create a new Quick List (point 1). Give it a name (point 2). Fill it with the commands we need (point 3). Go to Keymap, find Quick Lists, and then the adb we created and assign a key combination to it. In my case, it's . Let's press OK: Launch the application again and press and fast 4. Make sure that the application is closed. Option+F2 Option+F2 You can read more about the plugin here: https://plugins.jetbrains.com/plugin/7380-adb-idea https://github.com/pbreault/adb-idea Conclusion As you can see, a small ADB Idea plugin can speed up development considerably. You don't need to type long commands in the terminal and memorize them. You can set keyboard shortcuts on the commands and enter them into the . In my experience, there are a few disadvantages to ADB Idea: Quick list When updating the Android Studio, the plugin does not always have time to quickly update and is sometimes no longer compatible. Sometimes the plugin stops attaching to the application process. There is no possibility of creating custom more complex commands.
However, despite all these disadvantages, ADB Idea still greatly speeds up application development.

Accelerating Android Application Development With the ADB Idea Plugin

Acelerar el desarrollo de aplicaciones de Android con el complemento ADB Idea

Accélérer le développement d'applications Android avec le plugin ADB Idea

एडीबी आइडिया प्लगइन के साथ एंड्रॉइड एप्लिकेशन डेवलपमेंट को तेज करना

ADB Idea プラグインで Android アプリケーション開発を加速

Acelerando o desenvolvimento de aplicativos Android com o plug-in ADB Idea

Tăng tốc phát triển ứng dụng Android với plugin ADB Idea

使用 ADB Idea 插件加速 Android 应用程序开发

Authors:
(1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work;
(2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work;
(3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University;
(4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Table of Links Abstract and 1 Introduction 2 Calculating the Stochastic Wave Vector Dark Matter Signal 2.1 The Dark Photon Field 2.2 The Detector Signal 3 Statistical Analysis and 3.1 Signal Likelihood 3.2 Projected Exclusions 4 Application to Accelerometer Studies 4.1 Recasting Generalised Limits onto B − L Dark Matter 5 Future Directions 6 Conclusions, Acknowledgments, and References A Equipartition between Longitudinal and Transverse Modes B Derivation of Marginal Likelihood with Stochastic Field Amplitude C Covariance Matrix D The Case of the Gradient of a Scalar 4.1 Recasting Generalised Limits onto B − L Dark Matter To recast our limits on β shown in Fig. 4 into one on the parameter of interest for this model, the gauge coupling strength gB−L, we must define four quantities. Firstly, we must make clear what the quantity A is for this experiment, which will depend on both the model and the signal of interest for this sensor. Secondly, we must choose an observation time, Tobs. Thirdly, we must elaborate on what the concrete noise profile, σ(ω), for this type of experiment is. Lastly, and least importantly following from our discussion above, we must choose a latitude for the experiment. Once these are known, Eq. (2.8) can be re-arranged for gB−L, giving us our model- and experiment-specific 90% CL limit. For B − L coupled dark photon dark matter, the relevant signal is a differential acceleration. This is given by Eq. (2.6) in the time domain, with a now concrete choice for A, Here, gB−L is the gauge coupling strength of the model, ∆ij is the differential B −L charge per nucleon between materials i and j, given by For our observation time, we take Tobs = 10T⊕ to be firmly in the regime where the three peaks can be resolved. For light cavities, which may only be able to remain coherent over the scale of hours rather than days, such a runtime is optimistic. However, since our aim here is merely to showcase how our method can be used concretely and compare to the works of Refs. [36, 38], we do not see this as an issue. Our strategy is general and can be applied to any axial sensor and vector-like DM model, and we have settled on this configuration only for the sake of argument; other sensor technologies, such a magnetically levitated sensors, do not have this issue. Moreover, multiple cavities or data-stacking techniques can be employed to mitigate this issue. Our choices for all of the above parameters except for the laser power, which we expand on below, are summarised in Table 1. These are in keeping with the choices made in Refs. [36, 38]. The choice of where to tune the laser power is critical to give us competitive limits for a wide range of dark matter masses. We have found that we can achieve excellent limits for low dark matter masses, which are of most relevance to our work, by tuning the laser power such that the back-action and shot-noise components are minimised at low We note that this choice of power tuning is different from the strategies usually employed in other studies. Typically, the laser power is tuned so that the measurement-added noise is either minimised on resonance or, as LIGO implements it, well above resonance [88]. However, we found that both of these choices detriments the limit that we can draw at low masses, increasing the background at low frequencies beyond the thermal noise floor. For these other strategies to be beneficial at both the respective frequency targets and at low frequencies, the thermal noise would have to be significantly lower than that achieved with our choice of sensor mass, quality factor, and bath temperature. In the treatment of our backgrounds, we have neglected seismic noise, which can become important at frequencies below ∼ 10 Hz. Since we require the use of two materials to measure the differential acceleration peculiar to B − L DM, we can envision constructing two sensors: one with a moveable mirror made of material i and another of material j. By subtracting their signals, we both isolate the differential acceleration signature and remove backgrounds common to both—this includes the seismic noise component [63] We show our limits in Fig. 5 for three choices of resonance frequency: 1 Hz, 10 Hz and 100 Hz. For all but the last of these frequencies, we are able to exclude new regions of the UDLM B − L parameter space, best excluded by the fifth-force satellite experiment MICROSCOPE [89] and torsion-balance experiment E¨ot-Wash [90]. This showcases the power of such sensors in searching for this DM candidate, as was first pointed out in Ref. [63]. In a single-peak analysis focusing solely on the Compton peak, our limits would be weakened by approximately a factor of 2, as can be seen from Fig. 4. This difference becomes more dramatic for sensors located closer to the equator. We note that σ(ω), though mostly slowly varying in the frequency width ∆ω = ω⊕, does exhibit a large gradient around the resonance frequency of the cavity. In the neighbourhood of this frequency, our assumption that σ(ω) does not vary greatly in the above range is incorrect, and a more careful analysis in which all three peaks take on different noise levels would have to be conducted for more representative limits. However, we do not expect that our limits would differ greatly from our calculation and, at any rate, would still smoothly join to the regimes on either side of the resonance frequency, where our assumption holds. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work; (2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work; (3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University; (4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Authors: Authors: (1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work; (2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work; (3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University; (4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Calculating the Stochastic Wave Vector Dark Matter Signal 2.1 The Dark Photon Field 2.1 The Dark Photon Field 2.2 The Detector Signal 2.2 The Detector Signal 3 Statistical Analysis and 3.1 Signal Likelihood 3 Statistical Analysis and 3.1 Signal Likelihood 3.2 Projected Exclusions 3.2 Projected Exclusions 4 Application to Accelerometer Studies 4 Application to Accelerometer Studies 4.1 Recasting Generalised Limits onto B − L Dark Matter 4.1 Recasting Generalised Limits onto B − L Dark Matter 5 Future Directions 5 Future Directions 6 Conclusions, Acknowledgments, and References 6 Conclusions, Acknowledgments, and References A Equipartition between Longitudinal and Transverse Modes A Equipartition between Longitudinal and Transverse Modes B Derivation of Marginal Likelihood with Stochastic Field Amplitude B Derivation of Marginal Likelihood with Stochastic Field Amplitude C Covariance Matrix C Covariance Matrix D The Case of the Gradient of a Scalar D The Case of the Gradient of a Scalar 4.1 Recasting Generalised Limits onto B − L Dark Matter To recast our limits on β shown in Fig. 4 into one on the parameter of interest for this model, the gauge coupling strength gB−L, we must define four quantities. Firstly, we must make clear what the quantity A is for this experiment, which will depend on both the model and the signal of interest for this sensor. Secondly, we must choose an observation time, Tobs. Thirdly, we must elaborate on what the concrete noise profile, σ(ω), for this type of experiment is. Lastly, and least importantly following from our discussion above, we must choose a latitude for the experiment. Once these are known, Eq. (2.8) can be re-arranged for gB−L, giving us our model- and experiment-specific 90% CL limit. For B − L coupled dark photon dark matter, the relevant signal is a differential acceleration. This is given by Eq. (2.6) in the time domain, with a now concrete choice for A, Here, gB−L is the gauge coupling strength of the model, ∆ij is the differential B −L charge per nucleon between materials i and j, given by For our observation time, we take Tobs = 10T⊕ to be firmly in the regime where the three peaks can be resolved. For light cavities, which may only be able to remain coherent over the scale of hours rather than days, such a runtime is optimistic. However, since our aim here is merely to showcase how our method can be used concretely and compare to the works of Refs. [36, 38], we do not see this as an issue. Our strategy is general and can be applied to any axial sensor and vector-like DM model, and we have settled on this configuration only for the sake of argument; other sensor technologies, such a magnetically levitated sensors, do not have this issue. Moreover, multiple cavities or data-stacking techniques can be employed to mitigate this issue. Our choices for all of the above parameters except for the laser power, which we expand on below, are summarised in Table 1. These are in keeping with the choices made in Refs. [36, 38]. The choice of where to tune the laser power is critical to give us competitive limits for a wide range of dark matter masses. We have found that we can achieve excellent limits for low dark matter masses, which are of most relevance to our work, by tuning the laser power such that the back-action and shot-noise components are minimised at low We note that this choice of power tuning is different from the strategies usually employed in other studies. Typically, the laser power is tuned so that the measurement-added noise is either minimised on resonance or, as LIGO implements it, well above resonance [88]. However, we found that both of these choices detriments the limit that we can draw at low masses, increasing the background at low frequencies beyond the thermal noise floor. For these other strategies to be beneficial at both the respective frequency targets and at low frequencies, the thermal noise would have to be significantly lower than that achieved with our choice of sensor mass, quality factor, and bath temperature. In the treatment of our backgrounds, we have neglected seismic noise, which can become important at frequencies below ∼ 10 Hz. Since we require the use of two materials to measure the differential acceleration peculiar to B − L DM, we can envision constructing two sensors: one with a moveable mirror made of material i and another of material j. By subtracting their signals, we both isolate the differential acceleration signature and remove backgrounds common to both—this includes the seismic noise component [63] We show our limits in Fig. 5 for three choices of resonance frequency: 1 Hz, 10 Hz and 100 Hz. For all but the last of these frequencies, we are able to exclude new regions of the UDLM B − L parameter space, best excluded by the fifth-force satellite experiment MICROSCOPE [89] and torsion-balance experiment E¨ot-Wash [90]. This showcases the power of such sensors in searching for this DM candidate, as was first pointed out in Ref. [63]. In a single-peak analysis focusing solely on the Compton peak, our limits would be weakened by approximately a factor of 2, as can be seen from Fig. 4. This difference becomes more dramatic for sensors located closer to the equator. We note that σ(ω), though mostly slowly varying in the frequency width ∆ω = ω⊕, does exhibit a large gradient around the resonance frequency of the cavity. In the neighbourhood of this frequency, our assumption that σ(ω) does not vary greatly in the above range is incorrect, and a more careful analysis in which all three peaks take on different noise levels would have to be conducted for more representative limits. However, we do not expect that our limits would differ greatly from our calculation and, at any rate, would still smoothly join to the regimes on either side of the resonance frequency, where our assumption holds. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Recasting Three-Peak Exclusion Limits on B − L Dark Matter via Accelerometer Studies

Authors:
(1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work;
(2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work;
(3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University;
(4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Table of Links Abstract and 1 Introduction 2 Calculating the Stochastic Wave Vector Dark Matter Signal 2.1 The Dark Photon Field 2.2 The Detector Signal 3 Statistical Analysis and 3.1 Signal Likelihood 3.2 Projected Exclusions 4 Application to Accelerometer Studies 4.1 Recasting Generalised Limits onto B − L Dark Matter 5 Future Directions 6 Conclusions, Acknowledgments, and References A Equipartition between Longitudinal and Transverse Modes B Derivation of Marginal Likelihood with Stochastic Field Amplitude C Covariance Matrix D The Case of the Gradient of a Scalar 4 Application to Accelerometer Studies However, those works that performed a Fourier space analysis did so by only focusing on the Compton peak. As we discussed in Section 3, such an analysis for an axial sensor is sub-optimal, as it fails to capture the full signal and therefore suffers from a signal loss at a range of latitudes. Moreover, ignoring the randomness of the nuisance variables leads to an overly aggressive constraint, as was illustrated in Fig. 3. Our more holistic threepeak strategy, which also considers this stochasticity, retains the full signal and is largely latitude-independent in its constraining power. Therefore, this choice of sensor and DM model makes for an excellent case study with which to showcase the improved constraining power of our method. Furthermore, a proper vector treatment of the DM field in Fourier space, which includes the stochasticity of the ULDM field variables and the effect of the rotation of the Earth, has not been done. Ignoring the randomness of the nuisance variables leads to an overly aggressive constraint, as was also pointed out in Refs. [42, 79]. This paper is available on arxiv under CC BY 4.0 DEED license. [9] Here we assume that the sensitivity axis of the detector is locally fixed with respect to the zenith. Authors: (1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work; (2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work; (3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University; (4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Authors: Authors: (1) Dorian W. P. Amaral, Department of Physics and Astronomy, Rice University and These authors contributed approximately equally to this work; (2) Mudit Jain, Department of Physics and Astronomy, Rice University, Theoretical Particle Physics and Cosmology, King’s College London and These authors contributed approximately equally to this work; (3) Mustafa A. Amin, Department of Physics and Astronomy, Rice University; (4) Christopher Tunnell, Department of Physics and Astronomy, Rice University. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Calculating the Stochastic Wave Vector Dark Matter Signal 2.1 The Dark Photon Field 2.1 The Dark Photon Field 2.2 The Detector Signal 2.2 The Detector Signal 3 Statistical Analysis and 3.1 Signal Likelihood 3 Statistical Analysis and 3.1 Signal Likelihood 3.2 Projected Exclusions 3.2 Projected Exclusions 4 Application to Accelerometer Studies 4 Application to Accelerometer Studies 4.1 Recasting Generalised Limits onto B − L Dark Matter 4.1 Recasting Generalised Limits onto B − L Dark Matter 5 Future Directions 5 Future Directions 6 Conclusions, Acknowledgments, and References 6 Conclusions, Acknowledgments, and References A Equipartition between Longitudinal and Transverse Modes A Equipartition between Longitudinal and Transverse Modes B Derivation of Marginal Likelihood with Stochastic Field Amplitude B Derivation of Marginal Likelihood with Stochastic Field Amplitude C Covariance Matrix C Covariance Matrix D The Case of the Gradient of a Scalar D The Case of the Gradient of a Scalar 4 Application to Accelerometer Studies However, those works that performed a Fourier space analysis did so by only focusing on the Compton peak. As we discussed in Section 3, such an analysis for an axial sensor is sub-optimal, as it fails to capture the full signal and therefore suffers from a signal loss at a range of latitudes. Moreover, ignoring the randomness of the nuisance variables leads to an overly aggressive constraint, as was illustrated in Fig. 3. Our more holistic threepeak strategy, which also considers this stochasticity, retains the full signal and is largely latitude-independent in its constraining power. Therefore, this choice of sensor and DM model makes for an excellent case study with which to showcase the improved constraining power of our method. Furthermore, a proper vector treatment of the DM field in Fourier space, which includes the stochasticity of the ULDM field variables and the effect of the rotation of the Earth, has not been done. Ignoring the randomness of the nuisance variables leads to an overly aggressive constraint, as was also pointed out in Refs. [42, 79]. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv [9] Here we assume that the sensitivity axis of the detector is locally fixed with respect to the zenith.

Enhanced Vector Dark Matter Detection in Accelerometer Studies Through a Three-Peak Analysis

Stories, News & Mentions about ACELERON

Categories

Trending Topics

Stories, News & Mentions about ACELERON

Aceleron’s Stories on HackerNoon

Aceleron's News & Mentions

Categories

Trending Topics