Instance Discrimination
Instance Discrimination is a self-supervised learning framework that belongs to the class of contrastive representation learning methods, distinguished by its approach of treating each data sample as a unique category and training a model to discriminate between different instances of the same underlying data distribution through similarity in embedding space.
107
Publications
9.3K
Citations
436
Authors
125
Institutions
Publications per year
2017–2026
102
Parents
107
Unsupervised Feature Learning via Non-parametric Instance Discrimination
Zhirong Wu, Yuanjun Xiong, Stella X. Yu et al. · 2018 · 3.6K citations
Understanding the Behaviour of Contrastive Loss
Feng Wang, Huaping Liu · 2021 · 589 citations
Prototypical Contrastive Learning of Unsupervised Representations
Junnan Li, Pan Zhou, Caiming Xiong et al. · arXiv (Cornell University) · 2020 · 484 citations · Full text
1–5 of 107
436
8
h-index
6
h-index
4
h-index
4
h-index
3
h-index
1–5 of 436
125
University of Science and Technology of China
Hefei, China
8 h-index
27 publications
809 citations

International Computer Science Institute
Berkeley, United States
5 h-index
10 publications
7.5K citations
University of California, Berkeley
Berkeley, United States
5 h-index
13 publications
7.5K citations
Chinese University of Hong Kong
Hong Kong, Hong Kong
4 h-index
9 publications
7.4K citations
Tsinghua University
Beijing, China
4 h-index
13 publications
1.7K citations
Salesforce (United States)
San Francisco, United States
3 h-index
6 publications
2K citations
University of Technology Sydney
Sydney, Australia
3 h-index
6 publications
149 citations

Shanghai Jiao Tong University
Shanghai, China
3 h-index
7 publications
182 citations
1–8 of 125
Repository · 35 publications · 1.6K citations
20
h-index
Conference · 6 publications · 259 citations
6
h-index
Conference · 5 publications · 269 citations
5
h-index
Conference · 4 publications · 487 citations
4
h-index
TPAMI · 5 publications · 335 citations
4
h-index
1–5