Benchmark and analysis of 165 pretrained SSL models. Code for "Evaluating Self-Supervised Learning via Risk Decomposition".

Official Code for ICLR2022 Paper: Chaos is a Ladder: A New Theoretical Understanding of Contrastive Learning via Augmentation Overlap

NeurIPS 2020, Debiased Contrastive Learning

PyTorch code and models for the DINOv2 self-supervised learning method.

CVPR 2022, Robust Contrastive Learning against Noisy Views

[ICML2022] Contrastive Learning with Boosted Memorization

The repository for a thorough empirical evaluation of pre-trained vision model performance across different downstream policy learning methods.

solo-learn: a library of self-supervised methods for visual representation learning powered by Pytorch Lightning

Preventing Dimensional Collapse in Self-Supervised Learning via Orthogonality Regularization