arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.03511cs.LGstat.ML
收藏

基于域不变表示的分布偏移下的泛化估计

Estimating Generalization under Distribution Shifts via Domain-Invariant Representations

Massachusetts Institute of Technology(麻省理工学院)

Ching-Yao Chuang, Antonio Torralba, Stefanie Jegelka

摘要:

当机器学习模型部署在与训练分布不同的测试分布上时,它们可能表现不佳,但高估了它们的性能。在这项工作中,我们的目标是更好地估计模型的性能下分布偏移,没有监督。为此,我们使用一组域不变预测器作为未知的真实目标标签的代理。由于所得到的风险估计的误差取决于代理模型的目标风险,我们研究了域不变表示的泛化,并表明潜在表示的复杂性对目标风险有显着影响。从经验上讲,我们的方法(1)能够自调整领域自适应模型,(2)准确地估计给定模型的目标误差下分布偏移。其他应用包括模型选择,决定提前停止和错误检测。

英文摘要:

When machine learning models are deployed on a test distribution different from the training distribution, they can perform poorly, but overestimate their performance. In this work, we aim to better estimate a model's performance under distribution shift, without supervision. To do so, we use a set of domain-invariant predictors as a proxy for the unknown, true target labels. Since the error of the resulting risk estimate depends on the target risk of the proxy model, we study generalization of domain-invariant representations and show that the complexity of the latent representation has a significant influence on the target risk. Empirically, our approach (1) enables self-tuning of domain adaptation models, and (2) accurately estimates the target error of given models under distribution shift. Other applications include model selection, deciding early stopping and error detection.

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。

↑