TACLE: Task and Class-aware Exemplar-free Semi-supervised Class Incremental Learning

TL;DR

TACLE is a novel framework for exemplar-free semi-supervised class incremental learning that leverages pre-trained models, a task-adaptive threshold, and class-aware loss to improve learning from limited labeled and unlabeled data.

Contribution

It introduces a task-adaptive threshold and class-aware weighted loss for better utilization of unlabeled data in exemplar-free incremental learning.

Findings

Effective on CIFAR10, CIFAR100, and ImageNet-Subset100 datasets.

Performs well with imbalanced unlabeled data.

Works with only one labeled example per class.

Abstract

We propose a novel TACLE (TAsk and CLass-awarE) framework to address the relatively unexplored and challenging problem of exemplar-free semi-supervised class incremental learning. In this scenario, at each new task, the model has to learn new classes from both (few) labeled and unlabeled data without access to exemplars from previous classes. In addition to leveraging the capabilities of pre-trained models, TACLE proposes a novel task-adaptive threshold, thereby maximizing the utilization of the available unlabeled data as incremental learning progresses. Additionally, to enhance the performance of the under-represented classes within each task, we propose a class-aware weighted cross-entropy loss. We also exploit the unlabeled data for classifier alignment, which further enhances the model performance. Extensive experiments on benchmark datasets, namely CIFAR10, CIFAR100, and ImageNet-Subset100 demonstrate the effectiveness of the proposed TACLE framework. We further showcase its effectiveness when the unlabeled data is imbalanced and also for the extreme case of one labeled example per class.