Simone Severini

In this paper, Linyu Liu etal’s study the problem of uncertainty estimation and calibration for LLMs. We first formulate the uncertainty estimation problem for LLMs and then propose a supervised approach that takes advantage of the labeled datasets and estimates the uncertainty of the LLMs’ responses. Based on the formulation, they illustrate the difference between the uncertainty estimation for #LLMs and that for standard #ML models and explain why the hidden neurons of the LLMs may contain uncertainty information. They designed an approach which demonstrates the benefits of utilising hidden activations to enhance uncertainty estimation across various tasks and shows robust transferability in out-of-distribution settings. They distinguish the uncertainty estimation task from the uncertainty calibration task and show that a better uncertainty estimation mode leads to a better calibration performance. The method is easy to implement and adaptable to different levels of model accessibility including black box, grey box, and white box. The key is to develop an LLM which provides the user with a level of understanding of the likelihood that the results are good enough to make decisions, or not. https://lnkd.in/eYYHRZxz #Uncertainty #Estimation #LLMs #AI

9

Gravity might be a universal decoherence mechanism, underlying quantum-to-classical transitions and the emergence of classical behaviors from a fundamentally quantum universe. https://lnkd.in/dijVRFzV

2

Great thought provoking essay by AI veteran Rodney Brooks evaluating different AI applications. Various capability predictions in the areas of FSD, Robotics, LLM, DL and Space Flights are listed along with timelines. They are labeled with one of the following by Brooks 1)accurate 2)too pessimistic 3)too optimistic. He labels almost all the predictions to be too optimistic. It will be interesting to see how many people start calling Brooks a Luddite. He attributes unrealistic expectations of humans to 4 sins 1)Performances vs Competence We are pretty good at predicting capabilities of other humans based on our observation of current capabilities. But this doesn’t extend to other animals or AI systems 2)Indistinguishable from magic When people cannot explain how something works they cannot know its limits as they do not have any sort of model. They start attributing magical qualities to that something 3)Exponentialism The sin of exponentialism is that some other process that has a strong computational component is going to follow a Moore’s-like law when it is unwarranted to so argue. 4)Speed of deployment New technology development is much slower than what people expect #ai #prediction #llm #robot #fsd

17

2 Comments

Making quantum error correction work, moving forward in quantum roadmap Michael Newman and Kevin Satzinger, Research Scientists, Google Research, Google Quantum AI team Google introduce Willow, the first quantum processor where error-corrected qubits get exponentially better as they get bigger. Quantum computers offer promising applications in many fields, ranging from chemistry and drug discovery to optimization and cryptography. Most of these applications require billions if not trillions of operations to execute reliably — not much compared to what your web browser is doing right now. But quantum information is delicate, and even state-of-the-art quantum devices will typically experience at least one failure in every thousand operations. To achieve their potential, performance needs to improve dramatically. https://lnkd.in/df8v926y #google #googleresearch #googlequantumai #googlecloud #googlecloudconsulting

7

1 Comment

💡 ByteDance's Latest Work on Neural Network-Based Molecular Force Fields 💡 I recently read ByteDance’s work on developing a new force field, and it’s pretty good! 👍 ByteFF is a molecular mechanics force field trained on a massive dataset: 2.4 million geometries and 3.2 million torsion profiles. It leverages GNN to predict bonded and non-bonded parameters, showing strong performance in predicting molecular geometry, torsion energy, and conformational energy. This model has great potential for drug discovery and molecular simulation. I do have some concerns about using torsion scan + hessian methods to train non-bonded parameters in this framework. Non-bonded interactions often involve long-range effects, which are challenging for GNNs to capture accurately. That said, the overall results presented in this work are still quite promising. It's exciting to see tech giants like ByteDance stepping into AI for Science, and their vast resources enable them to support DFT calculations on this massive scale. (🤞 Hopefully, they will release the dataset for open use, LOL!) 👉 You can find the paper here: https://lnkd.in/gjmGV4fd 👉 And check out their open-source framework on GitHub, which is well-engineered: https://lnkd.in/gkxcTi6t #AIforScience #DrugDiscovery #MolecularModeling #ForceField #ByteDance

49

1 Comment

Finally - achieving and surpassing the benchmarks that have long been held to mark a tipping point in quantum computing. Today's announcement, taken together with other disclosure in the past few weeks, extends our leadership in the quantum computing sector and brings reliable quantum computing out from the lab into the real world. This achievement should be one that all of us in the sector can enjoy and appreciate. I look forward to building on these developments as we get to the point where it becomes impossible for classical computers to simulate quantum processors. A few weeks ago we demonstrated our ability to scale our quantum computers. Then we disclosed stunning results with Microsoft on logical qubit development. And today we provide details of our world leading quantum volume numbers (exponentially better than literally anyone else) and for the first time, 2 qubit gate fidelities above '3 nines' across all our zones in our fully connected quantum processor. I could not be prouder of my colleagues. There is so much more to come as we continue to set the high bar in the industry and lead by example in exemplifying excellence. https://lnkd.in/esMQfmNt

227

17 Comments

CNBC's Deirdre Bosa reports on Google's quantum chip development as Google unveils 'Willow', a new quantum computing chip. Google parent Alphabet shares jumped 6% up on the news! https://lnkd.in/gwrtQx6C

4

No time to waist. Start planning to get your data, applications, systems and devices Quantum Safe! Do you know the impact of not being PQ-Ready? Any idea how your cryptographic Bill Of Material (CBOM) looks like, and logical next steps to become quantum safe. How do you know your suppliers will be quantum safe? How do you measure? Ensure this topic will be soon on your next board meetings agenda! Get ready!

2

Ex-OpenAI Superalignment employee Leopold Aschenbrenner published a lengthy paper this week about how he thinks we could get to AGI (Artificial General Intelligence) by 2027. THAT IS SOON. It’s a long and complicated document but basically posits that between increases in compute, algorithmic improvements and a few ‘mysterious’ other things, we could actually speed this up. The implications are pretty significant overall and it’s worth diving into what he has to say even if you don’t agree with it. #openAI #technology #chatGPT #research #machinelearning #ai

2

https://lnkd.in/gGpqT8V4 - Chicago business media says the Illinois government is pressing hard to build a $20 billion, 150-acre quantum computing campus in Chicago. - The campus would be anchored by PsiQuantum. - The $20 billion figure refers to the potential for future business and government investments, as well as job creation. ‘Illinois may be on the verge of securing the largest technology project in its history—what is being labeled a “$20 billion, 150-acre quantum computing campus,” potentially anchored by Silicon Valley startup PsiQuantum, according to Crain’s Chicago Business. PsiQuantum, hot off an announcement that its receiving $600 million to build a manufacturing site in Australia, is reportedly considering two Chicago-area locations for the project, the business journal reports. ‘The proposed sites, the former U.S. Steel plant on the South Side and the former Texaco refinery in Lockport, are both under final review, with a decision expected soon. This initiative is part of a broader vision by Governor J.B. Pritzker’s administration, which pundits are referring to a modern-day Manhattan Project, to position Illinois as a leader quantum computing. ‘The initiative would initially create about 1,000 construction jobs and up to 500 direct jobs, including positions for hardware and software engineers. The project is designed to be accessible, with many positions requiring only associate’s degrees & specialized training. Additionally, the site would feature a cryogenic facility, essential for stabilizing subatomic particles used in quantum computing, which could cost between $300 million to $500 million. 'The selection of the former industrial sites is strategic, providing the necessary infrastructure for industrial-scale helium production, and the substantial water & power needed for the cryogenic process. The campus plans also include an R&D center to support the development of quantum [technology]. ‘The state’s investment in quantum technology is already considerable. Governor Pritzker’s administration previously allocated $200 million from Illinois’s capital budget to quantum initiatives, aiding in securing an additional $275 million in federal funding for local research projects. With another $500 million requested from the state legislature for further investment, Illinois is trying to make a push to become a federally recognized quantum hub, a designation that could bring up to $70 million in additional funding. 'The bet on quantum is not guaranteed. The Colorado quantum ecosystem is also making a play for that federal designation. Quantum computing, itself, is making profound progress toward practical uses, but it is uncertain as to exactly when that quantum practicality is achieved... There is also stiff competition within the quantum computing field with several quantum approaches being explored and exploited..'

15

2 Comments

Much has been said about many companies’ desire for more compute (and data) to train large foundation models. I think it’s under-appreciated that we also have nowhere near enough compute available for inference on foundation models. Years ago, when I was leading teams at Google, Baidu, and Stanford that focused on scaling up deep learning, many semiconductor makers, data center operators, and researchers asked me if AI would continue to make good use of more compute if they kept delivering it. For many desktop workloads, like running a web browser, a faster CPU doesn’t help much beyond a certain point. So do we really need faster and faster AI processors? Each time, I confidently replied “yes!” and encouraged them to keep scaling up compute. (Sometimes, I added half-jokingly that I had never met a machine learning engineer who felt like they had enough compute. 😀) Fortunately, this prediction has been right so far. However, beyond training, we are also far from exhausting the benefits of faster and higher volumes of inference. Today, a lot of LLM output is for human consumption. A human might read around 250 words per minute, which is around 6 tokens per second (250 words/min / (0.75 words/token) / (60 secs/min)). So it might seem there’s little value to generating tokens much faster than this. But in an agentic workflow, an LLM might be prompted repeatedly to reflect on and improve its output, use tools, plan and execute multiple steps, or implement multiple agents that collaborate. So, we might generate hundreds of thousands of tokens or more before showing any output to a user. This makes fast token generation very desirable and makes slower generation a bottleneck to taking better advantage of existing models. That’s why I’m excited about the work of companies like Groq, which can generate hundreds of tokens per second. Recently, SambaNova also showed it can hit hundreds of tokens per second. Incidentally, faster, cheaper token generation will also help make running evaluations (evals), which can be slow and expensive today since it involves iterating over many examples, more palatable. Fortunately, both training and inference are rapidly becoming cheaper. I spoke with Cathie Wood and Charles Roberts of the investment firm ARK, which is famous for its bullish predictions on tech. They estimate that AI training costs are falling 75% a year. If they are right, a foundation model that costs $100M to train this year might cost $25M to train next year. Further, they report that for “enterprise scale use cases, inference costs seem to be falling at an annual rate of ~86%, even faster than training costs.” I don’t know how accurate these specific predictions will turn out to be, but with progress in both semiconductors and algorithms, I do see training and inference costs falling rapidly. This will be good for application builders and help AI agentic workflows lift off! [Original text: https://lnkd.in/dJ9tVGh7 ]

2,895

80 Comments

https://lnkd.in/eXbCsAnD “He who controls the GPUs, controls the universe.” Access to GPUs, and in particular those made by Nvidia, the leading supplier, is vital for any company that wants to be taken seriously in #ai. Nvidia’s dominance has driven its market value to more than $2trn. 'GPUs do the computational heavy-lifting needed to train & operate large #ai models. Yet, oddly, this is not what they were designed for. The initials stand for “graphics processing unit”, because such chips were originally designed to process video-game graphics. It turned out that, fortunately for Nvidia, they could be repurposed for #ai workloads. 'Might it be better to design specialist ai chips from scratch? That is what many companies, small & large, are now doing in a bid to topple Nvidia. Dedicated ai chips promise to make building and running ai models faster, cheaper or both. Any firm that can mount a credible threat will have no shortage of customers, who dislike its lofty prices and limited supplies. 'GPUs contain several thousand processing engines, or “cores”, which let them run thousands of versions of the same simple task at the same time. Running #ai models similarly involves running lots of copies of the same task in parallel. Figuring out how to rewrite ai code to run on GPUs was one of the factors that triggered the current ai boom. 'Yet GPUs have their limitations, particularly when it comes to the speed with which data can be shuffled on and off them... When training very large #ai models, some GPu cores may be idle as much as half of the time. Andrew Feldman, the boss of Cerebras, a startup based in Sunnyvale, CA, likens it to the gridlock in a grocery store on the day before Thanksgiving. “Everybody’s in a queue, so there are blockages in the parking lot, there are blockages in the aisles, blockages at the checkout. That’s exactly what’s happening with a gpu.” 'Cerebras’s response is to put 900,000 cores, plus lots of memory, onto a single, enormous chip, to reduce the complexity of connecting up multiple chips & piping data between them. 'Groq, a startup from Mountain View, CA, is taking a different approach. Its ai chips, called language processing units (lpus), are optimised to run #llms particularly quickly. 'Yet another approach is that taken by MatX, also based in CA. GPUs contain features & circuitry that provide flexibility for graphics, but are not needed for llms. The firm is working on getting rid of such unnecessary cruft.

13

3 Comments

📀 Quantum Platform-as-a-Service has a new precedent — Quantinuum announces Nexus, a full-stack quantum computing platform available now, for beta users. Plus, a framework and architecture proposal for fault-tolerant bosonic qubits with discrete-variable ancilla, optical fibers for improved quantum data transfer, insights into optimal quantum emitter design for quantum networks, and unprecedented quantum simulation for pharmaceutical design. ⚆ Quantinuum launches Nexus, a platform that simplifies quantum workflow and manages resources across multiple quantum backends; offers advanced resource controls, data management, and secure sharing tools ⚆ The University of Oxford's newly-funded Quantum Computing Hub will bring together over 50 experts from 18 institutions to focus on improving quantum hardware, scalable networking, and algorithm development ⚆ University of Bath physicists design new specialty optical fibers with micro-structured cores for quantum computing data transfer; the fibers incorporate complex air pocket patterns to improve compatibility with single-photon sources, qubits, and active optical components ⚆ A framework for fault-tolerant quantum computation uses a hybrid system of bosonic qubits and noisy discrete-variable ancillae and incorporates bosonic error correction and advanced quantum control; numerical demonstrations and a presentation of hardware-efficient architecture suggest that current hardware has the potential to demonstrate fault-tolerant computation using the proposed system and error correction operations ⚆ UC Santa Barbara researchers develop models to improve photon-based quantum communication by addressing inefficiencies caused by atomic vibrations in photon emitters; techniques to improve emitter brightness and efficiency include careful material selection, atomic-level engineering, and coupling emitters to a photonic cavity ⚆ University of Melbourne research team uses Oak Ridge National Laboratory Frontier supercomputer to achieve the first quantum simulation of biological systems at a scale necessary to accurately model drug performance ⚆ Don't miss a single qubit -- follow me & subscribe to the free newsletter for your daily quantum computing news & research insights. #quantumcomputing #quantumtechnology #innovation #quantumscience #quantumphysics

75

Very insightful work by Anthropic’s interpretability team. And an amazing paper, with outstanding writing and figures. The idea is very simple: interpret LLMs by leveraging sparse autoencoders as surrogate models of the MLP of transformer blocks, which allow one to disambiguate the superposition of features captured by a single neuron. A simple idea, but a very careful and complex execution, as it is often the case in our line of work. The paper goes into many details and provide a large array of insights, although the gist of the implementation remains obfuscated due to the closed source nature of Claude. Too bad, because this is the kind of work that we need to better understand and eventually trust LLMs. This is demonstrated by the authors in the section ‘Influence on Behavior’, where they show that clamping some features to either high or low value during inference is “remarkably effective at modifying model outputs in specific, interpretable ways”. Hopefully this kind of work is going to be replicated and generalized to open-weights models, such that we have new ways to steer their behavior. https://lnkd.in/eVym7f_f #interpretability #xai #explainableai #steerableai #anthropic #claude #anthropic

3