CLIMB talk with Yisong Yue: Controlling the Structure of Inference and Learning in Neural Networks
Title: Controlling the Structure of Inference and Learning in Neural Networks
Speaker: Yisong Yue, Professor of Computing and Mathematical Sciences at the California Institute of Technology
Abstract:
This talk presents progress towards developing structure- or architecture-aware mathematical frameworks for reasoning about inference and learning in deep neural networks. By being architecture-aware, we can extract concrete analytical insights about training neural networks that are somewhat opaque from the perspective of standard tools such as Lipschitz constants. I will present two research thrusts along this direction. The first thrust is a majorize-minimize framework that develops a novel architecture-aware trust region for deep learning optimization, which we call “Deep Relative Trust”. The second thrust is a control-theoretic treatment of neural ODEs (and related architectures), leading to new algorithms that can enforce desirable properties such as stability, adversarial robustness, and forward invariance.
Bio:
Yisong Yue is a Professor of Computing and Mathematical Sciences at the California Institute of Technology. He was previously a research scientist at Disney Research. Before that, he was a postdoctoral researcher in the Machine Learning Department and the iLab at Carnegie Mellon University. He received a Ph.D. from Cornell University and a B.S. from the University of Illinois at Urbana-Champaign. Yisong recently spent a 2-year sabbatical in the autonomous driving industry. Yisong is also the Senior Program Chair of the ICLR 2024 (International Conference on Learning Representations). Yisong’s research interests are centered around machine learning, and in particular getting theory to work in practice. To that end, his research agenda spans both fundamental and applied pursuits, from novel learning-theoretic frameworks all the way to deployment in autonomous driving on public roads. His work has been recognized with multiple paper awards and nominations, including in robotics, computer vision, sports analytics, machine learning for health, and information retrieval. At Latitude AI, he works on machine learning approaches to behavior modeling and motion planning for autonomous driving.
CLIMB
The Center for the Theoretical Foundations of Learning, Inference, Information, Intelligence, Mathematics and Microeconomics at Berkeley (CLIMB) was established to address new conceptual and mathematical challenges arising at the interface between technology, science, and society. CLIMB recognizes the emergence of a new generation of technology that focuses on data, inferences, and decisions, a development that is leading to a deeper engagement between technology and real-world phenomena that involve human decisions, values, discoveries, and culture.
Website: www.climb.berkeley.edu
Title: Controlling the Structure of Inference and Learning in Neural Networks
Speaker: Yisong Yue, Professor of Computing and Mathematical Sciences at the California Institute of Technology
Abstract:
This talk presents progress towards developing structure- or architecture-aware mathematical frameworks for reasoning about inference and learning in deep neural networks. By being architecture-aware, we can extract concrete analytical insights about training neural networks that are somewhat opaque from the perspective of standard tools such as Lipschitz constants. I will present two research thrusts along this direction. The first thrust is a majorize-minimize framework that develops a novel architecture-aware trust region for deep learning optimization, which we call “Deep Relative Trust”. The second thrust is a control-theoretic treatment of neural ODEs (and related architectures), leading to new algorithms that can enforce desirable properties such as stability, adversarial robustness, and forward invariance.
Bio:
Yisong Yue is a Professor of Computing and Mathematical Sciences at the California Institute of Technology. He was previously a research scientist at Disney Research. Before that, he was a postdoctoral researcher in the Machine Learning Department and the iLab at Carnegie Mellon University. He received a Ph.D. from Cornell University and a B.S. from the University of Illinois at Urbana-Champaign. Yisong recently spent a 2-year sabbatical in the autonomous driving industry. Yisong is also the Senior Program Chair of the ICLR 2024 (International Conference on Learning Representations). Yisong’s research interests are centered around machine learning, and in particular getting theory to work in practice. To that end, his research agenda spans both fundamental and applied pursuits, from novel learning-theoretic frameworks all the way to deployment in autonomous driving on public roads. His work has been recognized with multiple paper awards and nominations, including in robotics, computer vision, sports analytics, machine learning for health, and information retrieval. At Latitude AI, he works on machine learning approaches to behavior modeling and motion planning for autonomous driving.
CLIMB
The Center for the Theoretical Foundations of Learning, Inference, Information, Intelligence, Mathematics and Microeconomics at Berkeley (CLIMB) was established to address new conceptual and mathematical challenges arising at the interface between technology, science, and society. CLIMB recognizes the emergence of a new generation of technology that focuses on data, inferences, and decisions, a development that is leading to a deeper engagement between technology and real-world phenomena that involve human decisions, values, discoveries, and culture.
Website: www.climb.berkeley.edu
