CODEs: Chamfer Out-of-Distribution Examples against Overconfidence Issue
This addresses the overconfidence issue in neural networks for OOD samples, which is critical for reliable AI in safety-critical domains, but it appears incremental as it builds on existing OOD handling methods.
The paper tackles the problem of overconfident predictions on out-of-distribution (OOD) samples in deep neural networks by proposing Chamfer OOD examples (CODEs), which are generated without extra data and used to suppress predictions, resulting in significant alleviation of the issue without harming classification accuracy and outperforming state-of-the-art methods.
Overconfident predictions on out-of-distribution (OOD) samples is a thorny issue for deep neural networks. The key to resolve the OOD overconfidence issue inherently is to build a subset of OOD samples and then suppress predictions on them. This paper proposes the Chamfer OOD examples (CODEs), whose distribution is close to that of in-distribution samples, and thus could be utilized to alleviate the OOD overconfidence issue effectively by suppressing predictions on them. To obtain CODEs, we first generate seed OOD examples via slicing&splicing operations on in-distribution samples from different categories, and then feed them to the Chamfer generative adversarial network for distribution transformation, without accessing to any extra data. Training with suppressing predictions on CODEs is validated to alleviate the OOD overconfidence issue largely without hurting classification accuracy, and outperform the state-of-the-art methods. Besides, we demonstrate CODEs are useful for improving OOD detection and classification.