no code implementations • 29 Feb 2024 • Fahimeh Hosseini Noohdani, Parsa Hosseini, Aryan Yazdan Parast, HamidReza Yaghoubi Araghi, Mahdieh Soleymani Baghshah
Based on our observations, models trained with ERM usually highly attend to either the causal components or the components having a high spurious correlation with the label (especially in datapoints on which models have a high confidence).
no code implementations • 8 Dec 2023 • Mahdi Ghaznavi, Hesam Asadollahzadeh, HamidReza Yaghoubi Araghi, Fahimeh Hosseini Noohdani, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah
In order to provide group robustness without such annotations, we propose a new method, called loss-based feature re-weighting (LFR), in which we infer a grouping of the data by evaluating an ERM-pre-trained model on a small left-out split of the training data.