Neural computations for geometric reasoning

Channel:
United States Simons Institute for the Theory of Computing
Subscribers:
68,700
Published on ● Video Link: https://www.youtube.com/watch?v=H_nltXkIPtw


Duration: 46:28
49 views
0

Bruno Olshausen (UC Berkeley)
https://simons.berkeley.edu/talks/bruno-olshausen-uc-berkeley-2024-06-03
Understanding Lower-Level Intelligence from AI, Psychology, and Neuroscience Perspectives

In order to survive in complex three-dimensional environments, animals must possess the ability to represent geometric structure and to perform computations on it. We have been exploring the neural computations that could enable such abilities. I will present recent progress on this problem, especially how such circuits could operate robustly at low power and small form factor.



Other Videos By Simons Institute for the Theory of Computing


2024-06-18Relational Predictive Reasoning in Brains and Machines
2024-06-18Physical problem-solving in minds and machines
2024-06-18Lightning Talks
2024-06-18A Multimodal Automated Interpretability Agent
2024-06-18Putting ourselves out of business
2024-06-18How does human and primate intelligence compare?
2024-06-18Empowerment Gain as Causal Learning
2024-06-18Lightning Talk
2024-06-18Seeing Faces in Things: A Model and Dataset for Pareidolia
2024-06-18Characterizing the impact of noise on neural computation
2024-06-18Neural computations for geometric reasoning
2024-06-13What is the right "token" for next-token visual prediction?
2024-06-13Combining innate concepts with learning to perceive the world
2024-06-13Toward a Theory of Perspective Perception in Pictures
2024-05-29Distribution Testing: Hypothesis Testing from Very Little (or Very Private) Data
2024-05-29Sublinear Time Algorithms for Estimating Edit Distance
2024-05-29Sketching and Randomized Linear Algebra
2024-05-29Testing Assumptions of Learning Algorithms
2024-05-29Adversarially Robust Streaming Algorithms
2024-05-29New Advances on Multi-Pass Graph Streaming Lower Bounds
2024-05-29Coresets and Their Applications to Design Sublinear Algorithms


Tags:
Simons Institute
theoretical computer science
UC Berkeley
Computer Science
Theory of Computation
Theory of Computing
Understanding Lower-Level Intelligence from AI; Psychology; and Neuroscience Perspectives
Bruno Olshausen