Counter-Current Learning: A Biologically Plausible Dual Network Approach for Deep Learning

TL;DR

This paper introduces counter-current learning (CCL), a biologically plausible dual network framework inspired by biological exchange mechanisms, enabling effective credit assignment and learning in neural networks without traditional backpropagation.

Contribution

The paper proposes a novel counter-current learning framework that employs dual networks with anti-parallel signaling, improving biological plausibility and performance in neural network training.

Findings

CCL achieves comparable accuracy to traditional algorithms on benchmark datasets.

The approach demonstrates effective unsupervised learning with autoencoders.

CCL offers a more biologically realistic alternative to backpropagation.

Abstract

Despite its widespread use in neural networks, error backpropagation has faced criticism for its lack of biological plausibility, suffering from issues such as the backward locking problem and the weight transport problem. These limitations have motivated researchers to explore more biologically plausible learning algorithms that could potentially shed light on how biological neural systems adapt and learn. Inspired by the counter-current exchange mechanisms observed in biological systems, we propose counter-current learning (CCL), a biologically plausible framework for credit assignment in neural networks. This framework employs a feedforward network to process input data and a feedback network to process targets, with each network enhancing the other through anti-parallel signal propagation. By leveraging the more informative signals from the bottom layer of the feedback network to guide the updates of the top layer of the feedforward network and vice versa, CCL enables the simultaneous transformation of source inputs to target outputs and the dynamic mutual influence of these transformations. Experimental results on MNIST, FashionMNIST, CIFAR10, and CIFAR100 datasets using multi-layer perceptrons and convolutional neural networks demonstrate that CCL achieves comparable performance to other biologically plausible algorithms while offering a more biologically realistic learning mechanism. Furthermore, we showcase the applicability of our approach to an autoencoder task, underscoring its potential for unsupervised representation learning. Our work presents a direction for biologically inspired and plausible learning algorithms, offering an alternative mechanism of learning and adaptation in neural networks.