Reducing Catastrophic Forgetting in Modular Neural Networks by Dynamic Information Balancing

TL;DR

This paper introduces a dynamic information balancing method for modular neural networks that reduces catastrophic forgetting in continual learning by adaptively routing inputs based on information load, improving performance over existing methods.

Contribution

It proposes a novel dynamic routing technique using reinforcement learning to balance information in modular neural networks, enhancing continual learning capabilities.

Findings

DIB combined with EWC outperforms similar models.

The method effectively reduces catastrophic forgetting.

Empirical results across multiple datasets support its effectiveness.

Abstract

Lifelong learning is a very important step toward realizing robust autonomous artificial agents. Neural networks are the main engine of deep learning, which is the current state-of-the-art technique in formulating adaptive artificial intelligent systems. However, neural networks suffer from catastrophic forgetting when stressed with the challenge of continual learning. We investigate how to exploit modular topology in neural networks in order to dynamically balance the information load between different modules by routing inputs based on the information content in each module so that information interference is minimized. Our dynamic information balancing (DIB) technique adapts a reinforcement learning technique to guide the routing of different inputs based on a reward signal derived from a measure of the information load in each module. Our empirical results show that DIB combined with elastic weight consolidation (EWC) regularization outperforms models with similar capacity and EWC regularization across different task formulations and datasets.