Skip to contentSkip to filters

We use cookies for analytics.

Towards Biologically Plausible Deep Learning: Early Inference in Energy-Based Models Approximates Back-Propagation

Asja Fischer

University of Bonn

Recording

35 min

Abstract

Asja Fischer describes how early Langevin inference steps in an energy-based latent-variable model approximate the propagation of error signals through hidden layers. In this account, changes in neural activity over time supply a learning signal for a local weight update. Theory and simulations connect the update to spike-timing-dependent plasticity and suggest a way for inference and credit assignment to share neural dynamics.

Topics