Local learning for stable backpropagation-free neural network training towards physical learning

TL;DR

This paper introduces FFzero, a local, forward-only learning framework enabling stable neural network training without backpropagation, suitable for physical neural networks and potentially reducing environmental impact.

Contribution

The paper presents FFzero, a novel local learning method that operates without backpropagation or automatic differentiation, applicable to multilayer perceptrons and convolutional neural networks.

Findings

FFzero enables stable training without backpropagation.

It generalizes to multilayer perceptrons and CNNs.

Simulations show viability for physical neural networks.

Abstract

While backpropagation and automatic differentiation have driven deep learning's success, the physical limits of chip manufacturing and rising environmental costs of deep learning motivate alternative learning paradigms such as physical neural networks. However, most existing physical neural networks still rely on digital computing for training, largely because backpropagation and automatic differentiation are difficult to realize in physical systems. We introduce FFzero, a forward-only learning framework enabling stable neural network training without backpropagation or automatic differentiation. FFzero combines layer-wise local learning, prototype-based representations, and directional-derivative-based optimization through forward evaluations only. We show that local learning is effective under forward-only optimization, where backpropagation fails. FFzero generalizes to multilayer perceptron and convolutional neural networks across classification and regression. Using a simulated photonic neural network as an example, we demonstrate that FFzero provides a viable path toward backpropagation-free in-situ physical learning.