polylog null 1
Country:
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.14)
- North America > United States (0.04)
- Europe > Germany (0.04)
- (4 more...)
Technology:
Country:
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.14)
- North America > United States (0.04)
- Europe > Germany (0.04)
- (4 more...)
Technology:
Contrastive Learning with Nasty Noise
Contrastive learning has emerged as a powerful paradigm for self-supervised representation learning. This work analyzes the theoretical limits of contrastive learning under nasty noise, where an adversary modifies or replaces training samples. Using PAC learning and VC-dimension analysis, lower and upper bounds on sample complexity in adversarial settings are established. Additionally, data-dependent sample complexity bounds based on the l2-distance function are derived.
2502.17872
Technology: Information Technology > Artificial Intelligence > Machine Learning > Computational Learning Theory (0.56)