Does Feedback Alignment Work at Biological Timescales?

TL;DR

This paper develops a continuous-time model for feedback alignment, revealing that learning depends on the temporal overlap of neural signals, and explains its robustness and limitations at biological timescales.

Contribution

It introduces a continuous-time framework for feedback alignment, linking its effectiveness to the temporal overlap of neural signals, and clarifies conditions for biological plausibility.

Findings

Learning depends on the overlap between presynaptic drive and error signals.

Moderate timing mismatch does not impair learning due to overlap robustness.

Complete mismatch eliminates overlap, causing learning failure.

Abstract

Feedback alignment and related weight-transport-free algorithms are often proposed as biologically plausible alternatives to backpropagation, yet they are typically formulated in discrete phases with implicitly synchronized forward and error signals. We develop a continuous-time model of feedback-alignment-type learning in which neural activities and synaptic weights evolve together under coupled first-order dynamics with distinct propagation, plasticity, and decay time constants. We show that learning is governed by the temporal overlap between presynaptic drive and a locally projected error signal, providing an analytic explanation for robustness to moderate timing mismatch and for failure when mismatch eliminates overlap. Our results show that in order for feedback-alignment-type algorithms to work at biological timescales, they must obey the same temporal overlap principle that applies to other biological processes like eligibility traces.