Seminar

Registration required to attend.

Making use of modern black-box AI tools such as deep reinforcement learning is potentially transformational for safety-critical systems such as data centers, the electricity grid, transportation, and beyond. However, such machine-learned algorithms typically do not have formal guarantees on their worst-case performance, stability, or safety. So while their performance may improve upon traditional approaches in “typical” cases, they may perform arbitrarily worse in scenarios where the training examples are not representative due to, for example, distribution shift. Thus, a challenging open question emerges: Is it possible to provide guarantees that allow black-box AI tools to be used in safety-critical applications? This talk will provide an overview of an emerging area in studying learning-augmented algorithms that seeks to answer this question in the affirmative. This talk will survey recent results in this area and describe applications of these results to the design of sustainable data centers and control of the smart grid. 

Adam Wierman is a professor in the Computing & Mathematical Sciences Department at Caltech. He received his PhD, MSc, and BSc in computer science from Carnegie Mellon University and has been a faculty member at Caltech since 2007. Wierman’s research strives to make the networked systems that govern our world sustainable and resilient. He is best known for his work spearheading the design of algorithms for sustainable data centers, and he is co-author of a recent book, The Fundamentals of Heavy Tails. He is the recipient of multiple awards, including the ACM SIGMETRICS Rising Star Researcher Award, the SIGMETRICS Test of Time Award, the IEEE Communications Society William R. Bennett Prize, and multiple teaching awards, and he is a co-author of papers that have received best paper awards at a wide variety of conferences across computer science, power engineering, and operations research.