Skip to contentSkip to filters

We use cookies for analytics.

How Much Intelligence Can You Get from Relaxation?

Alex Gower

University of Cambridge

Abstract

Learning in the brain, and in energy-efficient physical hardware, has to rely on local signals and continuous-time dynamics. This talk shows how networks that compute by relaxing on an energy landscape, from Hopfield-type networks to predictive coding, can be trained with purely local rules to shape the landscape itself. The resulting backpropagation-free generative models nearly match backpropagation on CIFAR-10 (3.64 vs 3.54 FID), which the speaker reports as the best quality yet reached without backpropagation. The talk then outlines how the same principle could let a system learn continually from a single stream of experience, such as a child's headcam video.

Topics