arXiv 2305.003742023-10-23cs.LGcs.CR

通过对抗性不变正则化增强对抗性对比学习

Enhancing Adversarial Contrastive Learning via Adversarial Invariant Regularization

National University of Singapore（新加坡国立大学）；RIKEN Center for Advanced Intelligence Project (AIP)（理化学研究所先进智能项目中心）；The University of Auckland（奥克兰大学）；The University of Melbourne（墨尔本大学）；The University of Tokyo（东京大学）

Xilie Xu, Jingfeng Zhang, Feng Liu, Masashi Sugiyama, Mohan Kankanhalli

备注：: NeurIPS 2023

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文利用因果推理解释对抗性对比学习（ACL），提出对抗性不变正则化（AIR）强制表示独立于风格因素，结合SIR显著提升了ACL在下游任务中的标准泛化性和鲁棒性。

摘要：

对抗性对比学习（ACL）是一种通过引入对抗性数据来增强标准对比学习（SCL）的技术，它能够学习到可抵御对抗性攻击和常见损坏的鲁棒表示，而无需昂贵的标注。为了提升可迁移性，现有工作引入了标准不变正则化（SIR），为SCL施加风格独立性属性，从而免除标准表示中混杂风格因素的影响。然而，风格独立性属性如何使ACL学习到的鲁棒表示受益尚不明确。在本文中，我们利用因果推理技术来解释ACL，并提出了对抗性不变正则化（AIR）以强制实现与风格因素的独立性。我们使用SIR和AIR共同调节ACL以输出鲁棒表示。在理论上，我们证明了AIR隐式地鼓励自然数据的不同视图与其对抗变体之间的表示距离独立于风格因素。在实验上，我们的结果表明，不变正则化在标准泛化性和下游任务的鲁棒性方面，显著提升了最先进ACL方法的性能。据我们所知，我们是首个应用因果推理来解释ACL并开发AIR以增强ACL学习到的鲁棒表示的团队。源代码位于 https://github.com/GodXuxilie/Enhancing_ACL_via_AIR。

英文摘要：

Adversarial contrastive learning (ACL) is a technique that enhances standard contrastive learning (SCL) by incorporating adversarial data to learn a robust representation that can withstand adversarial attacks and common corruptions without requiring costly annotations. To improve transferability, the existing work introduced the standard invariant regularization (SIR) to impose style-independence property to SCL, which can exempt the impact of nuisance style factors in the standard representation. However, it is unclear how the style-independence property benefits ACL-learned robust representations. In this paper, we leverage the technique of causal reasoning to interpret the ACL and propose adversarial invariant regularization (AIR) to enforce independence from style factors. We regulate the ACL using both SIR and AIR to output the robust representation. Theoretically, we show that AIR implicitly encourages the representational distance between different views of natural data and their adversarial variants to be independent of style factors. Empirically, our experimental results show that invariant regularization significantly improves the performance of state-of-the-art ACL methods in terms of both standard generalization and robustness on downstream tasks. To the best of our knowledge, we are the first to apply causal reasoning to interpret ACL and develop AIR for enhancing ACL-learned robust representations. Our source code is at https://github.com/GodXuxilie/Enhancing_ACL_via_AIR.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。