arXiv:2502.17872cs.LGcs.AI2025-02
研究对抗性噪声下对比学习的理论极限，揭示其样本复杂度边界。
Contrastive Learning with Nasty Noise
- 基于PAC学习与VC维分析，推导对抗环境下的样本复杂度上下界。
- 提出依赖数据的样本复杂度上界，以l2距离函数为衡量标准。
- 为鲁棒自监督学习提供理论支撑，适合关注模型安全性的研究者。
对比学习已成为自监督表示学习的强大范式。本文分析了在恶劣噪声（即攻击者修改或替换训练样本）条件下对比学习的理论极限。通过使用PAC学习和VC维分析，建立了对抗设置下的样本复杂度下界与上界。此外，还推导出基于l2距离函数的数据相关样本复杂度上界。
原文摘要 · Abstract (English)
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.
对比学习对抗噪声理论分析
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。