Papers
Topics
Authors
Recent
Search
2000 character limit reached

What If the Input is Expanded in OOD Detection?

Published 24 Oct 2024 in cs.LG | (2410.18472v2)

Abstract: Out-of-distribution (OOD) detection aims to identify OOD inputs from unknown classes, which is important for the reliable deployment of machine learning models in the open world. Various scoring functions are proposed to distinguish it from in-distribution (ID) data. However, existing methods generally focus on excavating the discriminative information from a single input, which implicitly limits its representation dimension. In this work, we introduce a novel perspective, i.e., employing different common corruptions on the input space, to expand that. We reveal an interesting phenomenon termed confidence mutation, where the confidence of OOD data can decrease significantly under the corruptions, while the ID data shows a higher confidence expectation considering the resistance of semantic features. Based on that, we formalize a new scoring method, namely, Confidence aVerage (CoVer), which can capture the dynamic differences by simply averaging the scores obtained from different corrupted inputs and the original ones, making the OOD and ID distributions more separable in detection tasks. Extensive experiments and analyses have been conducted to understand and verify the effectiveness of CoVer. The code is publicly available at: https://github.com/tmlr-group/CoVer.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (62)
  1. Extremely simple activation shaping for out-of-distribution detection. In ICLR, 2023.
  2. Feed two birds with one scone: Exploiting wild data for both out-of-distribution generalization and detection. In ICML, 2023.
  3. Breaking down out-of-distribution detection: Many methods based on OOD training data estimate a combination of the same core quantities. In ICML, 2022.
  4. In or out? fixing imagenet out-of-distribution detection evaluation. In ICML, 2023.
  5. On the opportunities and risks of foundation models. In arXiv, 2021.
  6. Altclip: Altering the language encoder in clip for extended language capabilities. In arXiv, 2022.
  7. Describing textures in the wild. In CVPR, 2014.
  8. Beyond mahalanobis distance for textual ood detection. In NeurIPS, 2022.
  9. A halfspace-mass depth-based method for adversarial attack detection. TMLR, 2023.
  10. Toward stronger textual attack detectors. In arXiv, 2023.
  11. Rainproof: An umbrella to shield text generators from out-of-distribution data. In arXiv, 2022.
  12. Unsupervised layer-wise score aggregation for textual ood detection. In AAAI, 2024.
  13. The relationship between precision-recall and ROC curves. In ICML, 2006.
  14. Imagenet: A large-scale hierarchical image database. In CVPR, 2009.
  15. An image is worth 16x16 words: Transformers for image recognition at scale. In arXiv, 2020.
  16. How does unlabeled data provably help out-of-distribution detection? In ICLR, 2024.
  17. Dream the impossible: Outlier imagination with diffusion models. In NeurIPS, 2023.
  18. VOS: learning what you don’t know by virtual outlier synthesis. In ICLR, 2022.
  19. Zero-shot out-of-distribution detection based on the pre-trained model clip. In AAAI, 2022.
  20. Exploring the limits of out-of-distribution detection. In NeurIPS, 2021.
  21. Deep learning. MIT Press, 2016.
  22. Scaling out-of-distribution detection for real-world settings. In ICML, 2022.
  23. Benchmarking neural network robustness to common corruptions and perturbations. In ICLR, 2019.
  24. A baseline for detecting misclassified and out-of-distribution examples in neural networks. In ICLR, 2017.
  25. On the importance of gradients for detecting distributional shifts in the wild. In NeurIPS, 2021.
  26. Dos: Diverse outlier sampling for out-of-distribution detection. In arXiv, 2023.
  27. Negative label guided ood detection with pretrained vision-language models. In ICLR, 2024.
  28. A tutorial on energy-based learning. In Predicting structured data, 2006.
  29. A simple unified framework for detecting out-of-distribution samples and adversarial attacks. In NeurIPS, 2018.
  30. Enhancing the reliability of out-of-distribution image detection in neural networks. In ICLR, 2018.
  31. Energy-based out-of-distribution detection. In NeurIPS, 2020.
  32. Delving into out-of-distribution detection with vision-language representations. In NeurIPS, 2022.
  33. Poem: Out-of-distribution detection with posterior sampling. In ICML, 2022.
  34. How does fine-tuning impact out-of-distribution detection for vision-language models? IJCV, 132(2):596–609, 2024.
  35. How to exploit hyperspherical embeddings for out-of-distribution detection? In arXiv, 2022.
  36. On the impact of spurious correlation for out-of-distribution detection. In AAAI, 2022.
  37. Locoop: Few-shot out-of-distribution detection via prompt learning. In NeurIPS, 2023.
  38. Reading digits in natural images with unsupervised feature learning. In NeurIPS Workshop, 2011.
  39. Deep neural networks are easily fooled: High confidence predictions for unrecognizable images. In CVPR, 2015.
  40. Nearest neighbor guidance for out-of-distribution detection. In CVPR, 2023.
  41. Adversarial attack detection under realistic constraints. 2022.
  42. A simple unsupervised data depth-based method to detect adversarial images. 2023.
  43. Learning transferable visual models from natural language supervision. In ICML, 2021.
  44. Barrage of random transforms for adversarially robust defense. In CVPR, 2019.
  45. Likelihood ratios for out-of-distribution detection. In NeurIPS, 2019.
  46. React: Out-of-distribution detection with rectified activations. In NeurIPS, 2021.
  47. Dice: Leveraging sparsification for out-of-distribution detection. In ECCV, 2022.
  48. Out-of-distribution detection with deep nearest neighbors. In ICML, 2022.
  49. Non-parametric outlier synthesis. In ICLR, 2023.
  50. The inaturalist species classification and detection dataset. In CVPR, 2018.
  51. Attention is all you need. In NeurIPS, 2017.
  52. Vim: Out-of-distribution with virtual-logit matching. In CVPR, 2022.
  53. Clipn for zero-shot ood detection: Teaching clip to say no. In ICCV, 2023.
  54. Learning to augment distributions for out-of-distribution detection. In NeurIPS, 2023.
  55. Watermarking for out-of-distribution detection. In NeurIPS, 2022.
  56. Sun database: Large-scale scene recognition from abbey to zoo. In CVPR, 2010.
  57. Demystifying clip data. In arXiv, 2023.
  58. Groupvit: Semantic segmentation emerges from text supervision. In CVPR, 2022.
  59. Openood: Benchmarking generalized out-of-distribution detection. In NeurIPS Datasets and Benchmarks Track, 2022.
  60. Generalized out-of-distribution detection: A survey. In arXiv, 2021.
  61. Places: A 10 million image database for scene recognition. TPAMI, 40(6):1452–1464, 2017.
  62. Diversified outlier exposure for out-of-distribution detection via informative extrapolation. In NeurIPS, 2023.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up