Effects of Auxiliary Knowledge on Continual Learning

TL;DR

This paper introduces a novel continual learning algorithm that leverages auxiliary data streams to enhance current task learning and facilitate future task adaptation, addressing the challenge of catastrophic forgetting.

Contribution

The proposed method uniquely combines main data with auxiliary, diverse streams to improve current task performance and future learning in continual learning settings.

Findings

Outperforms state-of-the-art models on CL image classification benchmarks.

Enhances feature discriminability and robustness through auxiliary data integration.

Facilitates knowledge transfer across tasks via auxiliary class mapping.

Abstract

In Continual Learning (CL), a neural network is trained on a stream of data whose distribution changes over time. In this context, the main problem is how to learn new information without forgetting old knowledge (i.e., Catastrophic Forgetting). Most existing CL approaches focus on finding solutions to preserve acquired knowledge, so working on the past of the model. However, we argue that as the model has to continually learn new tasks, it is also important to put focus on the present knowledge that could improve following tasks learning. In this paper we propose a new, simple, CL algorithm that focuses on solving the current task in a way that might facilitate the learning of the next ones. More specifically, our approach combines the main data stream with a secondary, diverse and uncorrelated stream, from which the network can draw auxiliary knowledge. This helps the model from different perspectives, since auxiliary data may contain useful features for the current and the next tasks and incoming task classes can be mapped onto auxiliary classes. Furthermore, the addition of data to the current task is implicitly making the classifier more robust as we are forcing the extraction of more discriminative features. Our method can outperform existing state-of-the-art models on the most common CL Image Classification benchmarks.