Elastic Representation: Mitigating Spurious Correlations for Group Robustness

TL;DR

This paper introduces Elastic Representation (ElRep), a method that reduces spurious correlations in deep learning models by applying norm penalties, thereby enhancing robustness across different groups without sacrificing overall accuracy.

Contribution

ElRep is a novel regularization technique that encourages feature diversity while mitigating spurious correlations, applicable to various deep learning models.

Findings

ElRep improves group robustness in neural networks.

ElRep minimally impacts in-distribution prediction performance.

Theoretical analysis confirms ElRep's balanced effect on overall accuracy.

Abstract

Deep learning models can suffer from severe performance degradation when relying on spurious correlations between input features and labels, making the models perform well on training data but have poor prediction accuracy for minority groups. This problem arises especially when training data are limited or imbalanced. While most prior work focuses on learning invariant features (with consistent correlations to y), it overlooks the potential harm of spurious correlations between features. We hereby propose Elastic Representation (ElRep) to learn features by imposing Nuclear- and Frobenius-norm penalties on the representation from the last layer of a neural network. Similar to the elastic net, ElRep enjoys the benefits of learning important features without losing feature diversity. The proposed method is simple yet effective. It can be integrated into many deep learning approaches to mitigate spurious correlations and improve group robustness. Moreover, we theoretically show that ElRep has minimum negative impacts on in-distribution predictions. This is a remarkable advantage over approaches that prioritize minority groups at the cost of overall performance.