mlsanityData to 7 Oct 2026

Resultant: Incremental Effectiveness on Likelihood for Unsupervised Out-of-Distribution Detection

arXiv 2409.03801 ↗First posted 5 Sep 2024stat.ML cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Unsupervised out-of-distribution (U-OOD) detection is to identify OOD data samples with a detector trained solely on unlabeled in-distribution (ID) data. The likelihood function estimated by a deep generative model (DGM) could be a natural detector, but its performance is limited in some popular "hard" benchmarks, such as FashionMNIST (ID) vs. MNIST (OOD). Recent studies have developed various detectors based on DGMs to move beyond likelihood. However, despite their success on "hard" benchmarks, most of them struggle to consistently surpass or match the performance of likelihood on some "non-hard" cases, such as SVHN (ID) vs. CIFAR10 (OOD) where likelihood could be a nearly perfect detector. Therefore, we appeal for more attention to incremental effectiveness on likelihood, i.e., whether a method could always surpass or at least match the performance of likelihood in U-OOD detection. We first investigate the likelihood of variational DGMs and find its detection performance could be improved in two directions: i) alleviating latent distribution mismatch, and ii) calibrating the dataset entropy-mutual integration. Then, we apply two techniques for each direction, specifically post-hoc prior and dataset entropy-mutual calibration. The final method, named Resultant, combines these two directions for better incremental effectiveness compared to either technique alone. Experimental results demonstrate that the Resultant could be a new state-of-the-art U-OOD detector while maintaining incremental effectiveness on likelihood in a wide range of tasks.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 68 of its 94 references that are in the map. First posted on arXiv 5 Sep 2024.

Builds on

68 of 94 references.

References · 68

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  7. 2022Generative Adversarial Networkscited by 5,603
  8. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  9. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  10. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  11. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  12. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
38 more
  1. 2013Describing Textures in the Wildcited by 2,110
  2. 2016Density estimation using Real NVPcited by 2,029
  3. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  4. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  5. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,428
  6. 2018Deep Anomaly Detection with Outlier Exposurecited by 939
  7. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  8. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
  9. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 515
  10. 2015Importance Weighted Autoencoderscited by 454
  11. 2019Likelihood Ratios for Out-of-Distribution Detectioncited by 451
  12. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  13. 2017PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modificationscited by 415
  14. 2018Deep Learning for Classical Japanese Literaturecited by 415
  15. 2021On the Importance of Gradients for Detecting Distributional Shifts in the Wildcited by 338
  16. 2022Understanding Diffusion Models: A Unified Perspectivecited by 337
  17. 2016Ladder Variational Autoencoderscited by 315
  18. 2021How to Train Your Energy-Based Modelscited by 271
  19. 2022Mitigating Neural Network Overconfidence with Logit Normalizationcited by 251
  20. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 220
  21. 2019Diagnosing and Enhancing VAE Modelscited by 192
  22. 2019Input complexity and out-of-distribution detection with likelihood-based generative modelscited by 186
  23. 2024Secrets of RLHF in Large Language Models Part II: Reward Modelingcited by 170
  24. 2020Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encodercited by 145
  25. 2020Learning Energy-Based Models by Diffusion Recovery Likelihoodcited by 122
  26. 2019Don't Blame the ELBO! A Linear VAE Perspective on Posterior Collapsecited by 120
  27. 2019BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modelingcited by 109
  28. 2022Denoising diffusion models for out-of-distribution detectioncited by 102
  29. 2021Understanding Failures in Out-of-Distribution Detection with Deep Generative Modelscited by 89
  30. 2022Training OOD Detectors in their Natural Habitatscited by 86
  31. 2021Provable Guarantees for Understanding Out-of-distribution Detectioncited by 78
  32. 2020Density of States Estimation for Out-of-Distribution Detectioncited by 74
  33. 2020Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Featurescited by 64
  34. 2021Hierarchical VAEs Know What They Don't Knowcited by 58
  35. 2021Open-set Label Noise Can Improve Robustness Against Inherent Label Noisecited by 58
  36. 2022Harnessing Out-Of-Distribution Examples via Augmenting Content and Stylecited by 43
  37. 2018Uncertainty in the Variational Information Bottleneckcited by 39
  38. 2023Learning to Augment Distributions for Out-of-Distribution Detectioncited by 38

50 most cited of 68 All 68 (JSON)

Built on by

Not cited yet.