Resilient Class-Incremental Learning: on the Interplay of Drifting, Unlabelled and Imbalanced Data Streams

TL;DR

This paper introduces SCIL, a resilient streaming class-incremental learning framework that effectively handles concept drift, class imbalance, and new class emergence in dynamic data streams, outperforming existing methods.

Contribution

The paper presents a novel SCIL framework combining autoencoders, dual-loss strategies, pseudo-label correction, and oversampling to improve incremental learning under challenging conditions.

Findings

SCIL outperforms state-of-the-art methods in experiments.

The dual-loss strategy enhances class detection and stability.

Oversampling and pseudo-label correction improve resilience to imbalance.

Abstract

In today's connected world, the generation of massive streaming data across diverse domains has become commonplace. In the presence of concept drift, class imbalance, label scarcity, and new class emergence, they jointly degrade representation stability, bias learning toward outdated distributions, and reduce the resilience and reliability of detection in dynamic environments. This paper proposes SCIL (Streaming Class-Incremental Learning) to address these challenges. The SCIL framework integrates an autoencoder (AE) with a multi-layer perceptron for multi-class prediction, uses a dual-loss strategy (classification and reconstruction) for prediction and new class detection, employs corrected pseudo-labels for online training, manages classes with queues, and applies oversampling to handle imbalance. The rationale behind the method's structure is elucidated through ablation studies and a comprehensive experimental evaluation is performed using both real-world and synthetic datasets that feature class imbalance, incremental classes, and concept drifts. Our results demonstrate that SCIL outperforms strong baselines and state-of-the-art methods. Based on our commitment to Open Science, we make our code and datasets available to the community.