Identifying Spurious Biases Early in Training through the Lens of Simplicity Bias

• Main Authors: Yang, Yu, Gan, Eric, Dziugaite, Gintare Karolina, Mirzasoleiman, Baharan
• Format: Journal Article
• Language: English
• Published: 30.05.2023
• Subjects: Computer Science - Computer Vision and Pattern Recognition; Computer Science - Learning
• DOI: 10.48550/arxiv.2305.18761

Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024, Valencia, Spain. PMLR: Volume 238 Neural networks trained with (stochastic) gradient descent have an inductive bias towards learning simpler solutions. This makes them highly prone to learning spurious correlations in the training data, that may not hold at test time. In this work, we provide the first theoretical analysis of the effect of simplicity bias on learning spurious correlations. Notably, we show that examples with spurious features are provably separable based on the model's output early in training. We further illustrate that if spurious features have a small enough noise-to-signal ratio, the network's output on the majority of examples is almost exclusively determined by the spurious features, leading to poor worst-group test accuracy. Finally, we propose SPARE, which identifies spurious correlations early in training and utilizes importance sampling to alleviate their effect. Empirically, we demonstrate that SPARE outperforms state-of-the-art methods by up to 21.1% in worst-group accuracy, while being up to 12x faster. We also show that SPARE is a highly effective but lightweight method to discover spurious correlations.