Robust Contrastive Learning

Robust contrastive learning aims to improve the reliability and robustness of self-supervised learning methods that learn representations from unlabeled data by comparing similar and dissimilar data points. Current research focuses on developing theoretically grounded loss functions, such as those based on Rényi divergence, and incorporating techniques like randomized smoothing and adversarial training to enhance resistance to noise and adversarial attacks. These advancements are significant because they improve the generalizability and reliability of contrastive learning models, leading to better performance in various applications, including image classification, medical image segmentation, and other domains with limited labeled data.

Papers

November 16, 2023

Robust Contrastive Learning With Theory Guarantee
Ngoc N. Tran, Lam Tran, Hoang Phan, Anh Bui, Tung Pham, Toan Tran, Dinh Phung, Trung Le
Contrastive Learning Theoretical Guarantee Self Supervised Loss Robust Loss Robust Contrastive Learning

August 31, 2023

Robust Representation Learning for Unreliable Partial Label Learning
Yu Shi, Dong-Dong Wu, Xin Geng, Min-Ling Zhang
Pseudo Label Partial Label Learning Robust Representation Learning Robust Contrastive Learning

February 5, 2023

Rethinking Robust Contrastive Learning from the Adversarial Perspective
Fatemeh Ghofrani, Mehdi Yaghouti, Pooyan Jamshidi
Contrastive Learning Adversarial Training Robust Deep Adversarial Perspective Robust Contrastive Learning

August 12, 2022

RenyiCL: Contrastive Representation Learning with Skew Renyi Divergence
Kyungmin Lee, Jinwoo Shin
Contrastive Learning Learned Representation Contrastive Representation Learning R\'enyi Divergence Robust Contrastive Learning

July 11, 2022

RUSH: Robust Contrastive Learning via Randomized Smoothing
Yijiang Pang, Boyang Liu, Jiayu Zhou
Adversarial Attack Adversarial Training Adversarial Robustness Randomized Smoothing Contrastive Pre Training Robust Contrastive Learning

July 5, 2022

MMGL: Multi-Scale Multi-View Global-Local Contrastive learning for Semi-supervised Cardiac Image Segmentation
Ziyuan Zhao, Jinxuan Hu, Zeng Zeng, Xulei Yang, Peisheng Qian, Bharadwaj Veeravalli, Cuntai Guan
Contrastive Learning Medical Image Segmentation Semi Supervised Medical Image Segmentation Robust Contrastive Learning

January 12, 2022

Robust Contrastive Learning against Noisy Views
Ching-Yao Chuang, R Devon Hjelm, Xin Wang, Vibhav Vineet, Neel Joshi, Antonio Torralba, Stefanie Jegelka, Yale Song
Contrastive Learning Synthesized View Contrastive Loss Function Symmetric Loss Robust Contrastive Learning