Improving Group-based Robustness and Calibration via Ordered Risk and Confidence Regularization

Neural network trained via empirical risk minimization achieves high accuracy on average but low accuracy on certain groups, especially when there is a spurious correlation. To construct the unbiased model from spurious correlation, we build a hypothesis that the inference to the samples without spurious correlation should take relative precedence over the inference to the spuriously biased samples. Based on the hypothesis, we propose the relative regularization to induce the training risk of each group to follow the specific order, which is sorted according to the degree of spurious correlation for each group. In addition, we introduce the ordering regularization based on the predictive confidence of each group to improve the model calibration, where other robust models still suffer from large calibration errors. These result in our complete algorithm, Ordered Risk and Confidence regularization (ORC). Our experiments demonstrate that ORC improves both the group robustness and calibration performances against the various types of spurious correlation in both synthetic and real-world datasets.