{
  "id": "2206.13687",
  "version": "v1",
  "published": "2022-06-28T01:35:43.000Z",
  "updated": "2022-06-28T01:35:43.000Z",
  "title": "POEM: Out-of-Distribution Detection with Posterior Sampling",
  "authors": [
    "Yifei Ming",
    "Ying Fan",
    "Yixuan Li"
  ],
  "comment": "ICML 2022 (Long Talk); First two authors contributed equally",
  "journal": "Thirty-ninth International Conference on Machine Learning (2022)",
  "categories": [
    "cs.LG",
    "cs.AI",
    "cs.CV"
  ],
  "abstract": "Out-of-distribution (OOD) detection is indispensable for machine learning models deployed in the open world. Recently, the use of an auxiliary outlier dataset during training (also known as outlier exposure) has shown promising performance. As the sample space for potential OOD data can be prohibitively large, sampling informative outliers is essential. In this work, we propose a novel posterior sampling-based outlier mining framework, POEM, which facilitates efficient use of outlier data and promotes learning a compact decision boundary between ID and OOD data for improved detection. We show that POEM establishes state-of-the-art performance on common benchmarks. Compared to the current best method that uses a greedy sampling strategy, POEM improves the relative performance by 42.0% and 24.2% (FPR95) on CIFAR-10 and CIFAR-100, respectively. We further provide theoretical insights on the effectiveness of POEM for OOD detection.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2022-06-28T01:35:43.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "out-of-distribution detection",
      "poem establishes state-of-the-art performance",
      "ood data",
      "current best method",
      "novel posterior sampling-based outlier"
    ],
    "tags": [
      "journal article"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}