Divide-and-Conquer for Enhancing Unlabeled Learning, Stability, and Plasticity in Semi-supervised Continual Learning

TL;DR

This paper introduces USP, a divide-and-conquer framework that significantly improves semi-supervised continual learning by enhancing unlabeled learning, stability, and plasticity through three novel strategies, leading to better performance over previous methods.

Contribution

The work presents a novel divide-and-conquer framework with three strategies—FSR, DCP, and CUD—that jointly improve unlabeled learning, stability, and plasticity in SSCL.

Findings

USP outperforms prior SSCL methods by up to 5.94% in accuracy.

The proposed strategies effectively balance learning plasticity and stability.

Comprehensive evaluations validate the framework's effectiveness.

Abstract

Semi-supervised continual learning (SSCL) seeks to leverage both labeled and unlabeled data in a sequential learning setup, aiming to reduce annotation costs while managing continual data arrival. SSCL introduces complex challenges, including ensuring effective unlabeled learning (UL), while balancing memory stability (MS) and learning plasticity (LP). Previous SSCL efforts have typically focused on isolated aspects of the three, while this work presents USP, a divide-and-conquer framework designed to synergistically enhance these three aspects: (1) Feature Space Reservation (FSR) strategy for LP, which constructs reserved feature locations for future classes by shaping old classes into an equiangular tight frame; (2) Divide-and-Conquer Pseudo-labeling (DCP) approach for UL, which assigns reliable pseudo-labels across both high- and low-confidence unlabeled data; and (3) Class-mean-anchored Unlabeled Distillation (CUD) for MS, which reuses DCP's outputs to anchor unlabeled data to stable class means for distillation to prevent forgetting. Comprehensive evaluations show USP outperforms prior SSCL methods, with gains up to 5.94% in the last accuracy, validating its effectiveness. The code is available at https://github.com/NJUyued/USP4SSCL.