Selective-Supervised Contrastive Learning with Noisy Labels

2022/03/08 by Shikun Li, Li, Shikun, Xiaobo Xia +5 · 13 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Music and Audio Processing #Speech and Audio Processing #cs.AI #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2203.04181

Accepted to CVPR 2022. 12 pages, 5 figure, and 10 tables

arxiv created 2022/03/08 · openalex publication_date 2022/03/08 · arxiv updated 2022/03/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Deep networks have strong capacities of embedding data into latent representations and finishing following tasks. However, the capacities largely come from high-quality annotated labels, which are expensive to collect. Noisy labels are more affordable, but result in corrupted representations, leading to poor generalization performance. To learn robust representations and handle noisy labels, we propose selective-supervised contrastive learning (Sel-CL) in this paper. Specifically, Sel-CL extend supervised contrastive learning (Sup-CL), which is powerful in representation learning, but is degraded when there are noisy labels. Sel-CL tackles the direct cause of the problem of Sup-CL. That is, as Sup-CL works in a pair-wise manner, noisy pairs built by noisy labels mislead representation learning. To alleviate the issue, we select confident pairs out of noisy ones for Sup-CL without knowing noise rates. In the selection process, by measuring the agreement between learned representations and given labels, we first identify confident examples that are exploited to build confident pairs. Then, the representation similarity distribution in the built confident pairs is exploited to identify more confident pairs out of noisy pairs. All obtained confident pairs are finally used for Sup-CL to enhance representations. Experiments on multiple noisy datasets demonstrate the robustness of the learned representations by our method, following the state-of-the-art performance. Source codes are available at https://github.com/ShikunLi/Sel-CL

Cited by

• Supervised Contrastive Frame Aggregation for Video Representation Learning
• Noise-robust Contrastive Learning for Critical Transition Detection in Dynamical Systems
• Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents
• Class Prototypes based Contrastive Learning for Classifying Multi-Label and Fine-Grained Educational Videos
• Flatness-Aware Stochastic Gradient Langevin Dynamics
• Robust Learning of Diffusion Models with Extremely Noisy Conditions
• Noisy-Pair Robust Representation Alignment for Positive-Unlabeled Learning
• Graph Your Own Prompt
• Reliable Programmatic Weak Supervision with Confidence Intervals for Label Probabilities
• Weakly-Supervised Contrastive Learning for Imprecise Class Labels
• Deformable Attentive Visual Enhancement for Referring Segmentation Using Vision-Language Model
• Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization
• EcoWikiRS: Learning Ecological Representation of Satellite Images from Weak Supervision with Species Observations and Wikipedia

Related

• Investigating Why Contrastive Learning Benefits Robustness Against Label Noise
• On Learning Contrastive Representations for Learning with Noisy Labels
• Channel-Wise Contrastive Learning for Learning with Noisy Labels
• An Inclusive Theoretical Framework of Robust Supervised Contrastive Loss against Label Noise
• Supervised Contrastive Learning