Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery

TL;DR

This paper demonstrates that large-scale self-supervised pre-trained models can effectively discover new classes in unlabeled data continuously and without supervision, outperforming more complex existing methods.

Contribution

It introduces a simple, robust baseline using a frozen pre-trained model and linear classifier for incremental novel class discovery without labeled data.

Findings

Pre-trained models outperform state-of-the-art methods in class discovery.

Simple baselines are effective and resilient in long-term learning scenarios.

Open-source code facilitates reproducibility and further research.

Abstract

Discovering novel concepts in unlabelled datasets and in a continuous manner is an important desideratum of lifelong learners. In the literature such problems have been partially addressed under very restricted settings, where novel classes are learned by jointly accessing a related labelled set (e.g., NCD) or by leveraging only a supervisedly pre-trained model (e.g., class-iNCD). In this work we challenge the status quo in class-iNCD and propose a learning paradigm where class discovery occurs continuously and truly unsupervisedly, without needing any related labelled set. In detail, we propose to exploit the richer priors from strong self-supervised pre-trained models (PTM). To this end, we propose simple baselines, composed of a frozen PTM backbone and a learnable linear classifier, that are not only simple to implement but also resilient under longer learning scenarios. We conduct extensive empirical evaluation on a multitude of benchmarks and show the effectiveness of our proposed baselines when compared with sophisticated state-of-the-art methods. The code is open source.