GAIT-prop: A biologically plausible learning rule derived from backpropagation of error

Abstract

Traditional backpropagation of error, though a highly successful algorithm for learning in artificial neural network models, includes features which are biologically implausible for learning in real neural circuits. An alternative called target propagation proposes to solve this implausibility by using a top-down model of neural activity to convert an error at the output of a neural network into layer-wise and plausible 'targets' for every unit. These targets can then be used to produce weight updates for network training. However, thus far, target propagation has been heuristically proposed without demonstrable equivalence to backpropagation. Here, we derive an exact correspondence between backpropagation and a modified form of target propagation (GAIT-prop) where the target is a small perturbation of the forward pass. Specifically, backpropagation and GAIT-prop give identical updates when synaptic weight matrices are orthogonal. In a series of simple computer vision experiments, we show near-identical performance between backpropagation and GAIT-prop with a soft orthogonality-inducing regularizer.

arXiv comment13 pages, 4 figures

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Deep Learning in Neural Networks: An Overview
• Deep Neural Networks Reveal a Gradient in the Complexity of Neural Representations across the Brain's Ventral Visual Pathway
• A sparse coding model with synaptically local plasticity and spiking neurons can account for the diverse shapes of V1 simple cell receptive fields
• How Auto-Encoders Could Provide Credit Assignment in Deep Networks via Target Propagation
• Deep Learning without Weight Transport
• Spike-based causal inference for weight alignment

Cited by in corpus (2)

• Deriving Differential Target Propagation from Iterating Approximate Inverses
• Cortico-cerebellar networks as decoupling neural interfaces