Credit Assignment in Neural Networks through Deep Feedback Control

TL;DR

Deep Feedback Control (DFC) is a biologically plausible learning method for neural networks that uses feedback control to assign credit, approximates Gauss-Newton optimization, and is supported by theoretical and experimental evidence.

Contribution

The paper introduces DFC, a novel local learning rule for neural networks that aligns with biological processes and optimizes network performance.

Findings

DFC approximates Gauss-Newton optimization.

DFC is fully local in space and time.

Experimental results on benchmarks validate DFC's effectiveness.

Abstract

The success of deep learning sparked interest in whether the brain learns by using similar techniques for assigning credit to each synaptic weight for its contribution to the network output. However, the majority of current attempts at biologically-plausible learning methods are either non-local in time, require highly specific connectivity motives, or have no clear link to any known mathematical optimization method. Here, we introduce Deep Feedback Control (DFC), a new learning method that uses a feedback controller to drive a deep neural network to match a desired output target and whose control signal can be used for credit assignment. The resulting learning rule is fully local in space and time and approximates Gauss-Newton optimization for a wide range of feedback connectivity patterns. To further underline its biological plausibility, we relate DFC to a multi-compartment model of cortical pyramidal neurons with a local voltage-dependent synaptic plasticity rule, consistent with recent theories of dendritic processing. By combining dynamical system theory with mathematical optimization theory, we provide a strong theoretical foundation for DFC that we corroborate with detailed results on toy experiments and standard computer-vision benchmarks.