ExMap: Leveraging Explainability Heatmaps for Unsupervised Group Robustness to Spurious Correlations

arXiv:2403.13870v1 Announce Type: cross
Abstract: Group robustness strategies aim to mitigate learned biases in deep learning models that arise from spurious correlations present in their training datasets. However, most existing methods rely on the access to the label distribution of the groups, which is time-consuming and expensive to obtain. As a result, unsupervised group robustness strategies are sought. Based on the insight that a trained model's classification strategies can be inferred accurately based on explainability heatmaps, we introduce ExMap, an …

arxiv correlations cs.cv cs.lg explainability robustness type unsupervised