Incremental False Negative Detection for Contrastive Learning

Abstract

Self-supervised learning has recently shown great potential in vision tasks through contrastive learning, which aims to discriminate each image, or instance, in the dataset. However, such instance-level learning ignores the semantic relationship among instances and sometimes undesirably repels the anchor from the semantically similar samples, termed as "false negatives". In this work, we show that the unfavorable effect from false negatives is more significant for the large-scale datasets with more semantic concepts. To address the issue, we propose a novel self-supervised contrastive learning framework that incrementally detects and explicitly removes the false negative samples. Specifically, following the training process, our method dynamically detects increasing high-quality false negatives considering that the encoder gradually improves and the embedding space becomes more semantically structural. Next, we discuss two strategies to explicitly remove the detected false negatives during contrastive learning. Extensive experiments show that our framework outperforms other self-supervised contrastive learning methods on multiple benchmarks in a limited resource setup.

arXiv commentICLR 2022

References in corpus (7)

• Ward's Hierarchical Clustering Method: Clustering Criterion and Agglomerative Algorithm
• Improved Baselines with Momentum Contrastive Learning
• Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
• Learning deep representations by mutual information estimation and maximization
• SGDR: Stochastic Gradient Descent with Warm Restarts
• Contrastive Multiview Coding
• A Theoretical Analysis of Contrastive Unsupervised Representation Learning

Cited by in corpus (3)

• Learning Representation for Clustering via Prototype Scattering and Positive Sampling
• Adaptive Soft Contrastive Learning
• Improving Contrastive Learning of Sentence Embeddings with Case-Augmented Positives and Retrieved Negatives