ARC 2021 Rick Stevens Abstract and Biography
AI for Precision Medicine: how close are we to changing the landscape?
Precision medicine methods identify subpopulations who differ in their disease risk, prognosis and response to treatment due to differences in underlying biology and other characteristics. Artificial intelligence (AI) leverages sophisticated computation and inference to generate insights, enables the system to reason and learn, and empowers clinician decision making through augmented intelligence. The convergence of AI and precision medicine promises to revolutionize health care. In this talk I'll review problems where machine learning (ML) and AI methods are making impact (antibiotic resistance, drug discovery, drug response prediction, molecular dynamics acceleration, etc.) in precision medicine. I'll discuss what we are learning about how precision medicine problems differ from more typical use cases for ML and what lessons we are learning from those working on very large problems (e.g. GPT-3). I'll also talk a bit about the "AI for Science" DOE initiative launched in 2019 and its relevance to develop a broad national program to accelerate scientific progress in biology, chemistry, materials science, cosmology, etc. One emerging concept I'll highlight, is the idea of "Autonomous Discovery" via Self-Driving Labs (SDL). Argonne, UChicago and other institutions are working on the concept of automating bench level science with (eventually) fully automated closed loop discovery systems, that can generate thousands of hypotheses, autonomously create the constructs, test them, update ML models, improve predicted outcomes and then loop, terminating when targets have been achieved or resources have been exhausted. Argonne and UChicago have launched an SDL initiative that is organizing teams across domains to develop the tools and infrastructure to bring AI to multiple areas of experimental discovery and to work these methods into our courses and graduate student training. Near term goals include large-scale AI driven design in energy materials for advanced batteries, improved throughput of constructs in synthetic biology, discovery of mixtures that promote polymer recycling, and protein design. Finally, I'll summarize our effort to develop a National AI Hardware Accelerator testbed that is being deployed in the Argonne Leadership Computing Facility that aims to provide researchers with access to systems from Cerebrase, SambaNova, Groq and GraphCore.
Rick Stevens, PhD, is Argonne’s Associate Laboratory Director for Computing, Environment and Life Sciences, and professor of Computer Science at the University of Chicago. Stevens has been at Argonne since 1982, and has served as director of the Mathematics and Computer Science Division and also as Acting Associate Laboratory Director for Physical, Biological and Computing Sciences. He is currently leader of Argonne’s Exascale Computing Initiative, and a Professor of Computer Science at the University of Chicago Physical Sciences Collegiate Division. From 2000-2004, Stevens served as Director of the National Science Foundation’s TeraGrid Project and from 1997-2001 as Chief Architect for the National Computational Science Alliance.
Stevens is interested in the development of innovative tools and techniques that enable computational scientists to solve important large-scale problems effectively on advanced scientific computers. Specifically, his research focuses on three principal areas: advanced collaboration and visualization environments, high-performance computer architectures (including Grids) and computational problems in the life sciences. In addition to his research work, Stevens teaches courses on computer architecture, collaboration technology, virtual reality, parallel computing and computational science.