Centroids Matching: an efficient Continual Learning approach operating in the embedding space

TL;DR

This paper introduces Centroids Matching, a novel regularization method for continual learning that operates in the embedding space, effectively reducing catastrophic forgetting with minimal memory use and faster performance.

Contribution

The paper presents a new regularization approach, Centroids Matching, that mitigates catastrophic forgetting by matching feature vectors to class centroids, requiring less memory and computation than existing methods.

Findings

Achieves high accuracy across multiple datasets and scenarios.

Operates efficiently with minimal memory footprint.

Outperforms existing methods in speed and effectiveness.

Abstract

Catastrophic forgetting (CF) occurs when a neural network loses the information previously learned while training on a set of samples from a different distribution, i.e., a new task. Existing approaches have achieved remarkable results in mitigating CF, especially in a scenario called task incremental learning. However, this scenario is not realistic, and limited work has been done to achieve good results on more realistic scenarios. In this paper, we propose a novel regularization method called Centroids Matching, that, inspired by meta-learning approaches, fights CF by operating in the feature space produced by the neural network, achieving good results while requiring a small memory footprint. Specifically, the approach classifies the samples directly using the feature vectors produced by the neural network, by matching those vectors with the centroids representing the classes from the current task, or all the tasks up to that point. Centroids Matching is faster than competing baselines, and it can be exploited to efficiently mitigate CF, by preserving the distances between the embedding space produced by the model when past tasks were over, and the one currently produced, leading to a method that achieves high accuracy on all the tasks, without using an external memory when operating on easy scenarios, or using a small one for more realistic ones. Extensive experiments demonstrate that Centroids Matching achieves accuracy gains on multiple datasets and scenarios.