Markus B. Huber

Put the power of quantum computing right into your workflows. Access everything you need: hybrid solvers, SDK, open-source tools, libraries, code examples, demos, and comprehensive documentation. It’s all available through the Leap™ quantum cloud service. Begin building quantum applications with ease today: https://lnkd.in/eNJUJMx

10

1 Comment

Rajesh Gupta spoke with SDSC director, Frank Würthwein, and others on the future of AI education at SC Conference Series #HPC #SDSC #SC24

4

Check out this highly recommended course for those interested in optimizing their models: - Introduction to Quantization - Symmetric and Asymmetric Quantization Modes - Hands-on Implementation from Scratch using PyTorch - covering Per Tensor, Per Channel (Axis), and Per Group techniques - Weights Packing: Exploring 2 or 4 Bits Quantization Methods https://lnkd.in/gMAuWXHm

43

What if we could design life-saving drug candidates atom by atom? At Pacific Northwest National Laboratory, we're turning this into reality by harnessing the power of #ArtificialIntelligence and #Computing, we've developed technology that integrates generative AI, reinforcement learning, and deep learning to advance #DrugDiscovery. This cutting-edge approach accelerates the identification of promising drug molecules and drastically reduces cost and time. Imagine swiftly crafting treatments for emerging diseases and pushing the boundaries of data science and computational biology!   Huge thanks to Tom Rickey for capturing the essence of our work in his recent highlight, your ability to translate complex science into engaging stories is priceless. I am proud of the incredible team and our collaborative ecosystem that fuels innovation. Together, we're not just imagining the future-we're engineering it. Highlight/article link: https://lnkd.in/gqCFB6GM Let's keep pushing the envelope in #AI, #data, #Computing, and #EnergyEfficiency to unlock new horizons in scientific discovery and beyond. #Innovation #AIatPNNL #Science #FutureOfMedicine #PNNLTechnology #AIforGood

54

2 Comments

Underlying all complex multiphysics simulations are even more complex mathematical algorithms that solve the equations describing movement of physical phenomena. Learn about the role of #LLNL researchers in optimizing the algorithms for faster computation: https://lnkd.in/gmVr6Pcu

69

Mixture-of-Agents as an Event-Driven System 🤖☎️ This paper by Wang et al. shows you how to ensemble smaller LLMs to create a system that can outperform state-of-the-art larger models. We’ve implemented this paper in a fully async, event-driven workflow system thanks to Ravi Theja Desetty - treat each “small LLM” as an event-driven step that can process incoming events and respond to it, independently and in parallel. ✅ Take full advantage of processing an entire batch of requests  ✅ Cleaner, readable code LlamaPack: https://lnkd.in/gRuqQGk2 Learn more about workflows: https://lnkd.in/gqWwz7Xs

311

9 Comments

#RENCI researchers Anirban Mandal + Cong W. are listed as authors on a paper being presented at 2 PM ET in Room B311 at #SC24. https://lnkd.in/guG7Np4E #LLM #anomalydetection #computationalworkflows #computationalefficiency

11

Our #icra2024 paper is now available from #ieee dl. We proposed a new Inverse Reinforcement Learning based method for hand trajectory prediction. Mukund Mitra Gyanig K. Partha Pratim Chakrabarti

18

2 Comments

Building and training large language models (LLMs) for scientific discovery comes with unique challenges, often requiring more resources than a single organization can provide. That’s why multi-institutional collaboration is key. The Trillion Parameter Consortium is an international partnership focused on advancing generative AI for science and engineering. The Argonne-led consortium is hosting a workshop at #SC24 today, bringing together researchers and innovators to discuss solutions for scaling LLMs for scientific discovery. If you attend the session, let us know what you learned in the comments! ⬇️

188

1 Comment

Great introduction to quantum hardware. Essentially, what is a difficult quantum computing problem to solve in hardware (noise, wiring...) becomes easier when the qubits have an optimal materials environment as they move through the circuits (and are shielded from whatever is happening elsewhere in the room/building/city). Google is well set-up for future scaling of circuitry and providing industry-ready platforms for quantum machine learning.

15

Check out this video exploring reinforcement learning using Isaac Lab to train hundreds of inverted pendulums, all in simulation. 💡

1

Although this epsiode is > 2 years old, and a lot of things typically happen over such a short period, It is still a timely discussion. Different people view causal inference from different lens, but i always valued Dr. Hill's views on causal inference research and teachings ever since she co-wrote the seminal book on Multilevel/Hierarchical Models with Dr. Gelman back in 2006. Her focus on integrating Bayesian framework with causal inference is particularly interesting, which she argues can naturally facilitate decisions on how to formula a generative model to handle the counterfactual world. Her contributions to causal inference research led to major developments in Bayesian non-parametric causal inference through BART (Bayesian Additive Regression Trees) which can potentially handle the treatment ignobility assumption inherently in one model, instead of building a model for treatment assignment and another for response surface.

4

🚗 Reinforcement learning isn’t just for teaching dogs new tricks, it's the backbone of self-driving cars and autonomous labs. Check out Meghan Heintz's latest article to see how RL is enabling breakthroughs in both fields. #ReinforcementLearning

6

Hiring machine learning engineers isn’t just about technical skills—it’s about creating AI systems that work in the real world. In our latest The Lucent Perspective episode, Ron Green, co-founder and CTO of KUNGFU.AI, explains why production-grade AI requires a unique blend of technical expertise, software engineering know-how, and cultural alignment. Key takeaways from Ron’s experience: - Look beyond AI skills—prioritize scalability, design, and reliability in your hires. - Software engineering best practices are non-negotiable for production-ready AI. - Humility and adaptability make a world of difference in team dynamics. - Generative AI is exciting, but domain-specific AI often delivers better ROI. Want to build an AI team that delivers real results? Start by listening to this episode. Listen now on: 🎧 Spotify: https://bit.ly/3CsugzA 🍏 Apple: https://apple.co/4etJgKD #TheLucentPerspective #Leadership #ScaleUp #TechRecruitment #AI

1

This research by Sagar Soni, A. C. Khunt, and A. H. Hasmani  discusses the Einstein-Cartan theory, which extends #generalrelativity to incorporate a torsion tensor in #spacetime. The tetrad formalism is used to study #fieldequations, and the energy-momentum tensor is determined for a Weyssenhoff fluid with #anisotropicmatter. Read more here: https://lnkd.in/eS6hra6s #wormhole #differentialforms #einsteincartantheory

8