提出无标签影响度方法，分析自监督模型学了什么。
Where Did Your Model Learn That? Label-free Influence for Self-supervised Learning
- 用数据增强稳定性衡量样本对模型的影响。
- 发现自监督模型受关键样本影响方式与有监督不同。
- 适用于检测重复数据、异常样本和公平性分析。
自监督学习（SSL）已革新大规模无标签数据的利用方式，但预训练数据与学习表征之间的内在关系仍不清晰。传统基于梯度的数据归因工具如影响函数依赖标签，不适用于无标签的SSL场景。本文提出Influence-SSL，一种专为自监督学习设计的无标签影响函数定义方法。该方法利用模型表征对数据增强的稳定性来识别能解释模型预测的关键训练样本。我们提供了理论基础与实证证据，验证Influence-SSL在分析预训练自监督模型中的有效性。分析显示，自监督模型对关键样本的响应模式与有监督模型存在显著差异。此外，我们在重复数据检测、异常值识别和公平性分析等任务中验证了Influence-SSL的有效性。代码已开源：https://github.com/cryptonymous9/Influence-SSL。
原文摘要 · Abstract (English)
Self-supervised learning (SSL) has revolutionized learning from large-scale unlabeled datasets, yet the intrinsic relationship between pretraining data and the learned representations remains poorly understood. Traditional supervised learning benefits from gradient-based data attribution tools like influence functions that measure the contribution of an individual data point to model predictions. However, existing definitions of influence rely on labels, making them unsuitable for SSL settings. We address this gap by introducing Influence-SSL, a novel and label-free approach for defining influence functions tailored to SSL. Our method harnesses the stability of learned representations against data augmentations to identify training examples that help explain model predictions. We provide both theoretical foundations and empirical evidence to show the utility of Influence-SSL in analyzing pre-trained SSL models. Our analysis reveals notable differences in how SSL models respond to influential data compared to supervised models. Finally, we validate the effectiveness of Influence-SSL through applications in duplicate detection, outlier identification and fairness analysis. Code is available at: \url{https://github.com/cryptonymous9/Influence-SSL}.
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。