Neuroscience-Inspired Memory Replay for Continual Learning: A Comparative Study of Predictive Coding and Backpropagation-Based Strategies

TL;DR

This paper compares predictive coding and backpropagation-based generative replay strategies for continual learning, demonstrating that biologically-inspired predictive coding offers improved memory retention and transfer efficiency in neural networks.

Contribution

It introduces a neuroscience-inspired generative replay framework and provides a comprehensive comparison showing the advantages of predictive coding over traditional methods.

Findings

Predictive coding-based replay improves retention by 15.3% on average.

Predictive coding maintains competitive transfer efficiency.

Biologically-inspired mechanisms enhance continual learning performance.

Abstract

Continual learning remains a fundamental challenge in artificial intelligence, with catastrophic forgetting posing a significant barrier to deploying neural networks in dynamic environments. Inspired by biological memory consolidation mechanisms, we propose a novel framework for generative replay that leverages predictive coding principles to mitigate forgetting. We present a comprehensive comparison between predictive coding-based and backpropagation-based generative replay strategies, evaluating their effectiveness on task retention and transfer efficiency across multiple benchmark datasets. Our experimental results demonstrate that predictive coding-based replay achieves superior retention performance (average 15.3% improvement) while maintaining competitive transfer efficiency, suggesting that biologically-inspired mechanisms can offer principled solutions to continual learning challenges. The proposed framework provides insights into the relationship between biological memory processes and artificial learning systems, opening new avenues for neuroscience-inspired AI research.