Professor (Dr.) M.K. BHANDARI’s Post

OS-COPILOT: TOWARDS GENERALIST COMPUTER AGENTS WITH SELF-IMPROVEMENT https://lnkd.in/dQBvUe5G Autonomous interaction with the computer has been a longstanding challenge with great potential, and the recent proliferation of large language models (LLMs) has markedly accelerated progress in building digital agents. However, most of these agents are designed to interact with a narrow domain, such as a specific software or website. This narrow focus constrains their applicability for general computer tasks. To this end, we introduce OS-Copilot, a framework to build generalist agents capable of interfacing with comprehensive elements in an operating system (OS), including the web, code terminals, files, multimedia, and various third-party applications. We use OS-Copilot to create FRIDAY, a self-improving embodied agent for automating general computer tasks. On GAIA, a general AI assistants benchmark, FRIDAY outperforms previous methods by 35%, showcasing strong generalization to unseen applications via accumulated skills from previous tasks. We also present numerical and quantitative evidence that FRIDAY learns to control and self-improve on Excel and Powerpoint with minimal supervision. Our OSCopilot framework and empirical findings provide infrastructure and insights for future research toward more capable and general-purpose computer agents. https://lnkd.in/d62WsnFW

🚀 𝐀 𝐌𝐚𝐣𝐨𝐫 𝐋𝐞𝐚𝐩 𝐢𝐧 𝐀𝐈 & 𝐌𝐋 𝐇𝐚𝐫𝐝𝐰𝐚𝐫𝐞 SiFive is at the forefront, unveiling their high-end RISC-V cores designed especially for AI and machine learning applications. What stands out? ✅ 𝑷𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 𝑷870: A significant jump in single-thread performance. Tailored not just for consumers, but with features added for automotive use as well. ✅ 𝑰𝒏𝒕𝒆𝒍𝒍𝒊𝒈𝒆𝒏𝒄𝒆 𝑿390: Quadrupled sustained data bandwidth and enhanced vector computation capabilities. Plus, the VCIX offers bespoke vector instructions and acceleration. These innovations are game-changers for consumer, automotive, and datacentre landscapes. With such strides, we're ushering in an era of more robust and tailored AI solutions. 🔗 Want to deep dive into the specifics? Check out the full article here: https://lnkd.in/gR-BqR_F #SiFive #RISCV #AIHardware #MachineLearning #PerformanceP870 #IntelligenceX390 #VectorCompute #ConsumerTech #AutomotiveTech #DatacentreEvolution #VCIX #AIIntegration #ComputeInnovation #OptimizedHardware #TechBreakthrough

It's all about AI in this conversation bettween Tiffany Janzen and Microsoft's Vivek Pradeep as they talk about Copilot+ PCs and more! 😁 Copilot+ PCs are the fastest, most intelligent Windows PCs ever with AI infused at every layer, starting with the world’s most powerful PC Neural Processing Units (NPUs) capable of delivering 40+ TOPS of compute. The new class of PCs is up to 20 times more powerful and up to 100 times as efficient for running AI workloads compared to traditional PCs. 🤯 Interested in learning more? Keep an eye out for the full video, releasing soon 👌 In the meantime, learn about Copilot+ PCs here:  https://lnkd.in/gvJ-smAz

#BOM #AI #genai 📢 With NVIDIA's #GTC2024 kicking off today, Generative AI will no doubt be front and center. And as the landscape of software development expands to include GenAI-generated code 👩💻, David suggests that we must address the #BOM and risk management implications: "Implementing an AI/ML BOM makes logical sense from a risk management standpoint because it provides visibility into the complex data and components, dealing with AI-generated code, but simultaneously complicates the processes engineers already manage." Read David's full article on our blog: https://lnkd.in/eHnPaamq

Nice read about another angle to AI and how it will effect software development and #productsecurity.

Retrieval Augmented Generation (RAG) is a cutting-edge method allowing Large Language Models (LLMs) to access enterprise data. With the ability to overcome one of the pressing limitations of LLMs, domain specificity, RAG enables enterprise to deploy purpose-built virtual assistants, chatbots, and agents. This innovation has tremendous implications for knowledge workers in most industries, including those in the intelligence field. Check out their answers here:

cafe-society.news already sits at the forefront of what machine learning work can be performed entirely on device. However, our device-only approach precludes landing garishly over-funded venture capital deals because device only means we cannot use "LLM" in the tech stack (no ChatGPT). Today, cafe-society.news looks forward to doing a spike on pockengine. It is a way to do LLM assisted machine learning on device. https://lnkd.in/g4bwGSMy. Frankly, our ML technology is already doing just fine and we think it can pencil better than chatGPT's approach selling $20 bills for $10. But, there's lots of creativity within the LLM community and we wish to be proven wrong.

With the end of the year in sight, it is also time for trend look-outs towards 2024 - 17 predictions in the field of AI for 2024 by NVIDIA to highlight 3 of them 👇 📍 𝐎𝐩𝐞𝐧-𝐬𝐨𝐮𝐫𝐜𝐞 𝐬𝐨𝐟𝐭𝐰𝐚𝐫𝐞 𝐥𝐞𝐚𝐝𝐬 𝐭𝐡𝐞 𝐜𝐡𝐚𝐫𝐠𝐞: Once companies combine these headstart models with private or real-time data, they can begin to see accelerated productivity and cost benefits across the organization. AI computing and software are set to become more accessible on virtually any platform, from cloud-based computing and AI model foundry services to the data center, edge and desktop. Manuvir Das 📍𝐓𝐚𝐫𝐠𝐞𝐭 𝐥𝐨𝐜𝐤 𝐨𝐧 𝐀𝐈 𝐬𝐚𝐟𝐞𝐭𝐲: Companies will heighten their focus on transparency and interpretability in AI systems — and use new tools and methodologies to shed light on the decision-making processes of complex AI models. As the generative AI ecosystem rallies around safety, anticipate AI technologies becoming more reliable, trustworthy and aligned with human values. Nikki P. 📍𝐅𝐢𝐬𝐡𝐢𝐧𝐠 𝐟𝐨𝐫 𝐋𝐋𝐌 𝐠𝐨𝐥𝐝 𝐢𝐧 𝐞𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞 𝐝𝐚𝐭𝐚 𝐥𝐚𝐤𝐞𝐬: There’s no shortage of statistics on how much information the average enterprise stores — it can be anywhere in the high hundreds of petabytes for large corporations. Yet many companies report that they’re mining less than half that information for actionable insights. Charlie Boyle #AI #artificialintelligence #GenAI #GenerativeAI #LLM #llms #Nvidia #Cloud #AIEthics #ResponsibleAI #ResponsibleTech #Datacenter

AI is changing the future of platform engineering. The table below outlines some potential use cases for LLMs in platform engineering that can change the game. You can read here all the analysis about how AI boosts Internal Developer Platform (IDP): https://lnkd.in/e7yCKMwN

The Perfect Solution for AI Data Centers! In the fast-paced world of AI technology, staying ahead of the curve is critical. That's why we're excited to introduce the StratusPower UPS, designed specifically to meet the needs of AI data centers. What makes the StratusPower UPS different? It has advanced computing capabilities with a multi-core trigonometric math unit and a control law accelerator. This UPS can handle parallel processing and supports IEEE 754 double-precision math. With more than 100 measurement points, the StratusPower UPS not only provides reliable power but also unlocks the potential of AI technologies in the data center. By efficiently collecting and managing data, StratusPower UPS optimizes the performance of your AI systems. StratusPower UPS - Empowering AI Evolution! Find out more: https://bit.ly/41nKL6S #AI #DataCenter #StratusPowerUPS #Innovation #EmpowerAI #Technology #DataManagement