Caltech Young Investigators Lecture
The transformative power of learning lies in automating the design of complex systems. Yet, it does not today incorporate requirements organically, leading to solutions prone to biases and unsafe behavior. To realize its autonomous engineering potential, we must therefore develop learning methods capable of satisfying statistical requirements beyond the training data. In this talk, I will show when and how it is possible to do so. I will define constrained learning by extending the classical PAC framework to show that constrained learning is not harder than unconstrained learning. I will also derive a practical learning rule that under mild conditions can tackle constrained learning tasks by solving only unconstrained ERM problems, a duality that holds despite non-convexity. I will illustrate through specific examples how these advances enable the data-driven design of trustworthy systems that adhere to fairness, robustness, resource, and safety specifications. These contributions suggest how to go beyond the current objective-centric learning paradigm towards a constraint-driven learning one. This talk assumes no prior knowledge of statistical learning or duality theory.
Luiz Chamon received the B.Sc. and M.Sc. degrees in electrical engineering from the University of São Paulo, Brazil, in 2011 and 2015 and the Ph.D. degree in electrical and systems engineering from the University of Pennsylvania in 2020. He is currently a postdoc of the Simons Institute of the University of California, Berkeley. In 2009, he was an undergraduate exchange student of the Masters in Acoustics of the École Centrale de Lyon, France, and an Assistant Instructor and Consultant on nondestructive testing at INSACAST Formation Continue. From 2010 to 2014, he was a Signal Processing Consultant on a project with EMBRAER. In 2018, he was recognized by the IEEE Signal Processing Society for his distinguished work for the editorial board of the IEEE Transactions on Signal Processing. He also received both the best student paper and the best paper awards at IEEE ICASSP 2020. His research interests include optimization, signal processing, machine learning, statistics, and control.
This talk is part of the Caltech Young Investigators Lecture Series, sponsored by the Division of Engineering and Applied Science.