Mitigating Simplicity Bias in Deep Learning for Improved OOD Generalization and Robustness
This addresses a fundamental issue in deep learning that affects model reliability and fairness, particularly in real-world applications, though it is an incremental improvement over existing methods.
The paper tackles the problem of simplicity bias in neural networks, which leads to poor out-of-distribution generalization, by proposing a framework that encourages the use of diverse features through regularization based on conditional mutual information, resulting in enhanced OOD generalization and improved robustness and fairness.
Neural networks (NNs) are known to exhibit simplicity bias where they tend to prefer learning 'simple' features over more 'complex' ones, even when the latter may be more informative. Simplicity bias can lead to the model making biased predictions which have poor out-of-distribution (OOD) generalization. To address this, we propose a framework that encourages the model to use a more diverse set of features to make predictions. We first train a simple model, and then regularize the conditional mutual information with respect to it to obtain the final model. We demonstrate the effectiveness of this framework in various problem settings and real-world applications, showing that it effectively addresses simplicity bias and leads to more features being used, enhances OOD generalization, and improves subgroup robustness and fairness. We complement these results with theoretical analyses of the effect of the regularization and its OOD generalization properties.