Fine-Grained Representation Learning via Multi-Level Contrastive Learning without Class Priors

TL;DR

This paper introduces Contrastive Disentangling, a novel unsupervised learning framework that captures fine-grained features without relying on class labels, using multi-level contrastive strategies and entropy regularization.

Contribution

It proposes a class-agnostic contrastive learning method with multi-level contrastive and entropy losses to learn rich, fine-grained representations without class priors.

Findings

Outperforms existing methods on CIFAR-10, CIFAR-100, STL-10, and ImageNet-10.

Effectively captures semantic and fine-grained features without class labels.

Demonstrates robustness in scenarios with ambiguous or unavailable class information.

Abstract

Recent advances in unsupervised representation learning often rely on knowing the number of classes to improve feature extraction and clustering. However, this assumption raises an important question: is the number of classes always necessary, and do class labels fully capture the fine-grained features within the data? In this paper, we propose Contrastive Disentangling (CD), a framework designed to learn representations without relying on class priors. CD leverages a multi-level contrastive learning strategy, integrating instance-level and feature-level contrastive losses with a normalized entropy loss to capture semantically rich and fine-grained representations. Specifically, (1) the instance-level contrastive loss separates feature representations across samples; (2) the feature-level contrastive loss promotes independence among feature heads; and (3) the normalized entropy loss ensures feature diversity and prevents feature collapse. Extensive experiments on CIFAR-10, CIFAR-100, STL-10, and ImageNet-10 demonstrate that CD outperforms existing methods in scenarios where class information is unavailable or ambiguous. The code is available at https://github.com/Hoper-J/Contrastive-Disentangling.