The paper proves statistical consistency of contrastive loss for retrieval via an AUC criterion and derives generalization bounds O(1/m + 1/sqrt(n)) supervised and O(1/sqrt(m) + 1/sqrt(n)) self-supervised that explain benefits of large negative sets.
arXiv preprint arXiv:2311.03881 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Statistical Consistency and Generalization of Contrastive Representation Learning
The paper proves statistical consistency of contrastive loss for retrieval via an AUC criterion and derives generalization bounds O(1/m + 1/sqrt(n)) supervised and O(1/sqrt(m) + 1/sqrt(n)) self-supervised that explain benefits of large negative sets.