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

by   Jianhong Bai, et al.

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.


Contrastive Learning with Boosted Memorization

Self-supervised learning has achieved a great success in the representat...

FEND: A Future Enhanced Distribution-Aware Contrastive Learning Framework for Long-tail Trajectory Prediction

Predicting the future trajectories of the traffic agents is a gordian te...

Supervised Contrastive Learning on Blended Images for Long-tailed Recognition

Real-world data often have a long-tailed distribution, where the number ...

Self-Damaging Contrastive Learning

The recent breakthrough achieved by contrastive learning accelerates the...

No Shifted Augmentations (NSA): compact distributions for robust self-supervised Anomaly Detection

Unsupervised Anomaly detection (AD) requires building a notion of normal...

WDiscOOD: Out-of-Distribution Detection via Whitened Linear Discriminative Analysis

Deep neural networks are susceptible to generating overconfident yet err...

ViM: Out-Of-Distribution with Virtual-logit Matching

Most of the existing Out-Of-Distribution (OOD) detection algorithms depe...

Please sign up or login with your details

Forgot password? Click here to reset