Proceedings

Prototype-based Optimal Transport for Out-of-Distribution Detection

Ao Ke, Wenlong Chen, Chuanwen Feng, Yukun Cao, Xike Xie, S. Kevin Zhou, Lei Feng

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 3018-3026.

https://doi.org/10.24963/ijcai.2025/336

Keywords

  • Machine Learning: ML: Trustworthy machine learning
  • Data Mining: DM: Anomaly/outlier detection

Abstract

Detecting Out-of-Distribution (OOD) inputs is crucial for improving the reliability of deep neural networks in the real-world deployment. In this paper, inspired by the inherent distribution shift between in-distribution (ID) and OOD data, we propose a novel method that leverages optimal transport to measure the distribution discrepancy between test inputs and ID prototypes. The resulting transport costs are used to quantify the individual contribution of each test input to the overall discrepancy, serving as a desirable measure for OOD detection. To address the issue that solely relying on the transport costs to ID prototypes is inadequate for identifying OOD inputs closer to ID data, we generate virtual outliers to approximate the OOD region via linear extrapolation. By combining the transport costs to ID prototypes with the costs to virtual outliers, the detection of OOD data near ID data is emphasized, thereby enhancing the distinction between ID and OOD inputs. Extensive evaluations demonstrate the superiority of our method over state-of-the-art methods.

Ao Ke, Wenlong Chen, Chuanwen Feng, Yukun Cao, Xike Xie, S. Kevin Zhou, and Lei Feng. 2025. Prototype-based Optimal Transport for Out-of-Distribution Detection. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 3018–3026. https://doi.org/10.24963/ijcai.2025/336
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards