When Covariate-shifted Data Augmentation Increases Test Error And How to Fix It
Empirically, data augmentation sometimes improves and sometimes hurts test error, even when only adding points with labels from the true conditional distribution that the hypothesis class is expressive enough to fit. In this paper, we provide precise conditions under which data augmentation hurts test accuracy for minimum norm estimators in linear regression. To mitigate the failure modes of augmentation, we introduce X-regularization, which uses unlabeled data to regularize the parameters towards the non-augmented estimate. We prove that our new estimator never hurts test error and exhibits significant improvements over adversarial data augmentation on CIFAR-10.
Code (0)
등록된 구현이 없습니다.
Tasks
Data AugmentationregressionSimilar Papers 제목 키워드 기반
Rescuing referral failures during automated diagnosis of domain-shifted medical images
The success of deep learning models deployed in the real world depends critically on their ability to generalize well across diverse data domains. Here, we address a fundamental challenge with selective classification du…
DiagnosticDisease PredictionDomain GeneralizationFrom Local Geometry to Global Pseudo Labeling for Robust Positive Unlabeled Learning under Covariate Shift
Detecting covariate shift is critical for building reliable vision systems. While most prior work focuses on improving robustness to shift, explicitly detecting covariate shift remains underexplored. Existing approaches …
Learning from Label Proportions and Covariate-shifted Instances
In many applications, especially due to lack of supervision or privacy concerns, the training data is grouped into bags of instances (feature-vectors) and for each bag we have only an aggregate label derived from the ins…
Domain AdaptationEvaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift
Covariate shift has been shown to sharply degrade both predictive accuracy and the calibration of uncertainty estimates for deep learning models. This is worrying, because covariate shift is prevalent in a wide range of …
Data AugmentationFully Test-time Adaptation for Tabular Data
Tabular data plays a vital role in various real-world scenarios and finds extensive applications. Although recent deep tabular models have shown remarkable success, they still struggle to handle data distribution shifts,…
Data AugmentationTest-time Adaptation