Reactive Model Correction: Mitigating Harm to Task-Relevant Features via Conditional Bias Suppression

arXiv:2404.09601v1 Announce Type: new
Abstract: Deep Neural Networks are prone to learning and relying on spurious correlations in the training data, which, for high-risk applications, can have fatal consequences. Various approaches to suppress model reliance on harmful features have been proposed that can be applied post-hoc without additional training. Whereas those methods can be applied with efficiency, they also tend to harm model performance by globally shifting the distribution of latent features. To mitigate unintended overcorrection of model behavior, we …

abstract applications arxiv bias consequences correlations cs.ai cs.cv cs.lg data features harm networks neural networks reliance risk training training data type via