Continual Semi-Supervised Learning through Contrastive Interpolation Consistency

TL;DR

This paper introduces a continual semi-supervised learning method that effectively leverages limited labeled data and unlabeled data to prevent forgetting and outperform fully supervised models in continual learning scenarios.

Contribution

It proposes a novel CSSL approach using metric learning and consistency regularization, demonstrating improved resilience to limited supervision in continual learning.

Findings

Our method outperforms state-of-the-art supervised continual learning methods with only 25% labeled data.

The approach maintains high performance despite reduced supervision, reducing labeling costs.

Experiments show increased robustness to forgetting and overfitting in semi-supervised continual learning.

Abstract

Continual Learning (CL) investigates how to train Deep Networks on a stream of tasks without incurring forgetting. CL settings proposed in literature assume that every incoming example is paired with ground-truth annotations. However, this clashes with many real-world applications: gathering labeled data, which is in itself tedious and expensive, becomes infeasible when data flow as a stream. This work explores Continual Semi-Supervised Learning (CSSL): here, only a small fraction of labeled input examples are shown to the learner. We assess how current CL methods (e.g.: EWC, LwF, iCaRL, ER, GDumb, DER) perform in this novel and challenging scenario, where overfitting entangles forgetting. Subsequently, we design a novel CSSL method that exploits metric learning and consistency regularization to leverage unlabeled examples while learning. We show that our proposal exhibits higher resilience to diminishing supervision and, even more surprisingly, relying only on 25% supervision suffices to outperform SOTA methods trained under full supervision.