Event Details:
"Physics of intelligence" for trustworthy and green AI
Abstract:
Neuroscience is experiencing accelerating advances in the scale and resolution of neural activity recordings from animals engaged in natural behaviors. At the same time, in machine learning, large-scale artificial neural network models are engaging in natural conversations and generating realistic videos. In this talk, I will show how we can take advantage of these technological advances and derive fundamental principles of neural learning and computation by integrating scientific methods from physics, neuroscience, and machine learning. Furthermore, this understanding then allows us to invent new algorithms that make AI models more reliable and efficient. I present three key contributions: (i) a new framework for studying compositional generalization in multimodal generative models, (ii) a generalization of Noether's theorem in physics to explain how symmetry constrains the geometry of neural learning dynamics, and (iii) decoding the neural code of biological vision through interpretable AI.
Hidenori Tanaka
Harvard University
Hidenori Tanaka received his bachelor's degree in physics from Kyoto University, Japan. He then received his Ph.D. in theoretical condensed matter physics from Harvard University, where he was advised by David Nelson and Michael Brenner. He then switched to theoretical neuroscience and machine learning with Surya Ganguli and Daniel Fisher at Stanford. He is currently a Group Leader at the NTT Physics & Informatics Lab, where he physically leads the Intelligent Systems Group at Harvard's Center for Brain Science through an industry-academic collaboration.
About the Wu Tsai Neuro MBCT Seminar Series
The Stanford Center for Mind, Brain, Computation and Technology Seminars (MBCT) explores ways in which computational and technical approaches are being used to advance the frontiers of neuroscience. It features speakers from other institutions, Stanford faculty and senior training program trainees.
The MBCT Seminar Series is only offered in person.