On the Effectiveness of Out-of-Distribution Data in Self-Supervised Long-Tail Learning

TL;DR

This paper introduces COLT, a novel SSL method that leverages out-of-distribution data to improve long-tail learning, effectively re-balancing feature space and outperforming methods relying on expensive in-domain data.

Contribution

The paper proposes a new SSL approach that uses OOD data with a dynamic sampling strategy and a contrastive loss to enhance long-tail learning without requiring large-scale in-domain data.

Findings

Significant performance improvements on long-tailed datasets.

Outperforms methods using external in-domain data.

Effective OOD sampling strategy enhances SSL robustness.

Abstract

Though Self-supervised learning (SSL) has been widely studied as a promising technique for representation learning, it doesn't generalize well on long-tailed datasets due to the majority classes dominating the feature space. Recent work shows that the long-tailed learning performance could be boosted by sampling extra in-domain (ID) data for self-supervised training, however, large-scale ID data which can rebalance the minority classes are expensive to collect. In this paper, we propose an alternative but easy-to-use and effective solution, Contrastive with Out-of-distribution (OOD) data for Long-Tail learning (COLT), which can effectively exploit OOD data to dynamically re-balance the feature space. We empirically identify the counter-intuitive usefulness of OOD samples in SSL long-tailed learning and principally design a novel SSL method. Concretely, we first localize the `head' and `tail' samples by assigning a tailness score to each OOD sample based on its neighborhoods in the feature space. Then, we propose an online OOD sampling strategy to dynamically re-balance the feature space. Finally, we enforce the model to be capable of distinguishing ID and OOD samples by a distribution-level supervised contrastive loss. Extensive experiments are conducted on various datasets and several state-of-the-art SSL frameworks to verify the effectiveness of the proposed method. The results show that our method significantly improves the performance of SSL on long-tailed datasets by a large margin, and even outperforms previous work which uses external ID data. Our code is available at https://github.com/JianhongBai/COLT.