Séminaire DIC-ISC-CRIA - 5 février 2026 par Rajesh P. N. RAO

Rajesh P. N. RAO - 5 février 2026 à 10h30 au PK-5115 (201, ave President-Kennedy, 5e étage)

TITRE : Predictive coding and generative models in natural and artificial intelligence

RÉSUMÉ 

This talk explores how predictive coding principles illuminate the computational foundations of both natural and artificial intelligence. Rao will examine his recent work on Dynamic Predictive Coding and Active Predictive Coding (APC) models, which proposes that the brain uses hierarchical generative models to predict sensory inputs and motor consequences. The discussion will cover how these models enable compositionality, hierarchical learning, and efficient planning by combining perception and action in a unified framework. Neuroscience evidence and AI applications suggests how predictive coding can help  us understand biological intelligence and develop more capable artificial systems that learn hierarchical world models for perception, action, and cognition.

BIOGRAPHIE

Rajesh P. N. RAO is the CJ and Elizabeth Hwang Professor of Computer Science & Engineering and Electrical & Computer Engineering at the University of Washington, Seattle. He is co-Director of the Center for Neurotechnology and directs the Neural Systems Laboratory. Rao received his PhD from University of Rochester (1998) and was a Sloan Postdoctoral Fellow at the Salk Institute. His research spans computational neuroscience, brain-computer interfaces, and artificial intelligence. He co-proposed the predictive coding model of brain function with Dana Ballard in 1999. His awards include a Guggenheim Fellowship, IEEE Fellow award, Fulbright Scholar award, NSF CAREER award, ONR Young Investigator Award, Sloan Faculty Fellowship, and Packard Fellowship.

RÉFÉRENCES:

Jiang, L. P., & Rao, R. P. N. (2024). Dynamic predictive coding: A model of hierarchical sequence learning and prediction in the neocortex. PLOS Computational Biology, 20(2), e1011801.

Rao, R. P. N. (2024). Active Predictive Coding: A Unifying Neural Model for Active Perception, Compositional Learning, and Hierarchical Planning. Neural Computation, 36(1), 1-58.
Gklezakos, D. C., & Rao, R. P. N. (2024). A sensory-motor theory of the neocortex based on active predictive coding. Nature Neuroscience.

Rao, R. P. N., & Ballard, D. H. (1999). Predictive coding in the visual cortex: A functional interpretation of some extra-classical receptive-field effects. Nature Neuroscience, 2(1), 79-87.