Supervision Accelerates Pre-training in Contrastive Semi-Supervised Learning of Visual Representations.

By Mahmoud Assran, Nicolas Ballas, Lluís Castrejón, Michael G. Rabbat. 2020

18 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Joelle Pineau
2. Lucas Caccia
3. Timothée Lesort
4. Oleksiy Ostapenko
5. Daniel J. Lizotte
6. Liyuan Wang
7. Jianfeng Gao
8. Sergey Levine

Citations in the corpus

• A soft nearest-neighbor framework for continual semi-supervised learning (2022, neutral): “For instance, self-supervised pre-training was found beneficial in [11], exponential moving average normalization was adopted in [11] and contrastive learning was considered in [3].”
• SPeCiaL: Self-Supervised Pretraining for Continual Learning (2021, result): “This observation is consistent with [Assran et al. , 2020], who note that when combining supervised and SSL losses, after a fixed number of epochs the supervised signal needs to be removed to avoid a decrease in performance.”