Overcoming Catastrophic Forgetting by XAI

TL;DR

This paper introduces CFD, a tool for explaining catastrophic forgetting in neural networks, and proposes Critical Freezing, a new continual learning method that improves performance and enhances explainability.

Contribution

The work presents CFD for interpreting forgetting mechanisms and introduces Critical Freezing, a novel approach that outperforms existing methods in continual learning.

Findings

CFD reveals which network components forget during training.

Critical Freezing significantly reduces catastrophic forgetting.

The proposed method outperforms recent continual learning techniques.

Abstract

Explaining the behaviors of deep neural networks, usually considered as black boxes, is critical especially when they are now being adopted over diverse aspects of human life. Taking the advantages of interpretable machine learning (interpretable ML), this work proposes a novel tool called Catastrophic Forgetting Dissector (or CFD) to explain catastrophic forgetting in continual learning settings. We also introduce a new method called Critical Freezing based on the observations of our tool. Experiments on ResNet articulate how catastrophic forgetting happens, particularly showing which components of this famous network are forgetting. Our new continual learning algorithm defeats various recent techniques by a significant margin, proving the capability of the investigation. Critical freezing not only attacks catastrophic forgetting but also exposes explainability.