Shapley Neuron Values for Continual Learning: Which Neurons Matter Most?

TL;DR

This paper introduces Shapley Neuron Values (SNV), a novel framework based on cooperative game theory that identifies important neurons to mitigate catastrophic forgetting in continual learning without increasing model size.

Contribution

SNV is a new principled method for quantifying neuron importance, enabling buffer-free continual learning by selectively freezing critical neurons.

Findings

SNV outperforms existing buffer-free methods on ImageNet-1k.

SNV improves accuracy by +2.88% in class incremental learning.

SNV improves accuracy by +6.46% in task incremental learning.

Abstract

Continual learning enables neural networks to learn tasks sequentially without forgetting previously acquired knowledge. However, neural networks suffer from catastrophic forgetting, where learning new tasks degrades performance on earlier ones. We address this problem with Shapley Neuron Valuation (SNV), a principled framework that quantifies Neuron importance in continual learning, grounded in cooperative game theory. SNV selectively freezes important Neurons while keeping others plastic, enabling buffer-free continual learning without expanding architecture. Experiments on ImageNet-1k show that SNV consistently outperforms existing buffer-free methods. In particular, SNV improves accuracy by +2.88% in the class incremental learning and +6.46% in the task incremental learning scenarios compared to the second baseline.