Class-Incremental Experience Replay for Continual Learning under Concept Drift

TL;DR

This paper introduces a novel continual learning method that combines experience replay with concept drift detection, enabling models to remember, forget, and adapt to changing data distributions over time.

Contribution

It presents a unified architecture integrating centroid-driven experience replay and reactive subspace buffers to handle both knowledge retention and concept drift adaptation.

Findings

Effective in balancing learning new classes and forgetting outdated ones.

Handles concept drift by adapting clusters in the memory buffer.

Outperforms existing methods in dynamic data scenarios.

Abstract

Modern machine learning systems need to be able to cope with constantly arriving and changing data. Two main areas of research dealing with such scenarios are continual learning and data stream mining. Continual learning focuses on accumulating knowledge and avoiding forgetting, assuming information once learned should be stored. Data stream mining focuses on adaptation to concept drift and discarding outdated information, assuming that only the most recent data is relevant. While these two areas are mainly being developed in separation, they offer complementary views on the problem of learning from dynamic data. There is a need for unifying them, by offering architectures capable of both learning and storing new information, as well as revisiting and adapting to changes in previously seen concepts. We propose a novel continual learning approach that can handle both tasks. Our experience replay method is fueled by a centroid-driven memory storing diverse instances of incrementally arriving classes. This is enhanced with a reactive subspace buffer that tracks concept drift occurrences in previously seen classes and adapts clusters accordingly. The proposed architecture is thus capable of both remembering valid and forgetting outdated information, offering a holistic framework for continual learning under concept drift.