Improving Out-of-Distribution Robustness via Selective Augmentation

01/02/2022
by   Huaxiu Yao, et al.
109

Machine learning algorithms typically assume that training and test examples are drawn from the same distribution. However, distribution shift is a common problem in real-world applications and can cause models to perform dramatically worse at test time. In this paper, we specifically consider the problems of domain shifts and subpopulation shifts (eg. imbalanced data). While prior works often seek to explicitly regularize internal representations and predictors of the model to be domain invariant, we instead aim to regularize the whole function without restricting the model's internal representations. This leads to a simple mixup-based technique which learns invariant functions via selective augmentation called LISA. LISA selectively interpolates samples either with the same labels but different domains or with the same domain but different labels. We analyze a linear setting and theoretically show how LISA leads to a smaller worst-group error. Empirically, we study the effectiveness of LISA on nine benchmarks ranging from subpopulation shifts to domain shifts, and we find that LISA consistently outperforms other state-of-the-art methods.

READ FULL TEXT
research
10/25/2022

Multi-Domain Long-Tailed Learning by Augmenting Disentangled Representations

There is an inescapable long-tailed class-imbalance issue in many real-w...
research
09/19/2022

Importance Tempering: Group Robustness for Overparameterized Models

Although overparameterized models have shown their success on many machi...
research
05/13/2021

Causally-motivated Shortcut Removal Using Auxiliary Labels

Robustness to certain distribution shifts is a key requirement in many M...
research
07/12/2023

Single Domain Generalization via Normalised Cross-correlation Based Convolutions

Deep learning techniques often perform poorly in the presence of domain ...
research
06/05/2022

AugLoss: A Learning Methodology for Real-World Dataset Corruption

Deep Learning (DL) models achieve great successes in many domains. Howev...
research
02/23/2023

Change is Hard: A Closer Look at Subpopulation Shift

Machine learning models often perform poorly on subgroups that are under...
research
06/04/2022

Toward Learning Robust and Invariant Representations with Alignment Regularization and Data Augmentation

Data augmentation has been proven to be an effective technique for devel...

Please sign up or login with your details

Forgot password? Click here to reset