MaskTune: Mitigating Spurious Correlations by Forcing to Explore
This addresses the issue of spurious correlations in over-parameterized models for machine learning practitioners, offering an incremental improvement by building on existing mitigation approaches without extra annotations.
The paper tackles the problem of deep learning models overfitting to spurious features by proposing MaskTune, a masking strategy that forces models to explore new features during finetuning without requiring additional supervision. It shows effectiveness on biased datasets like MNIST and CelebA, achieving similar or better performance than competing methods in selective classification tasks.
A fundamental challenge of over-parameterized deep learning models is learning meaningful data representations that yield good performance on a downstream task without over-fitting spurious input features. This work proposes MaskTune, a masking strategy that prevents over-reliance on spurious (or a limited number of) features. MaskTune forces the trained model to explore new features during a single epoch finetuning by masking previously discovered features. MaskTune, unlike earlier approaches for mitigating shortcut learning, does not require any supervision, such as annotating spurious features or labels for subgroup samples in a dataset. Our empirical results on biased MNIST, CelebA, Waterbirds, and ImagenNet-9L datasets show that MaskTune is effective on tasks that often suffer from the existence of spurious correlations. Finally, we show that MaskTune outperforms or achieves similar performance to the competing methods when applied to the selective classification (classification with rejection option) task. Code for MaskTune is available at https://github.com/aliasgharkhani/Masktune.