mlsanityData to 14 Oct 2026

Anomaly Detection with Variance Stabilized Density Estimation

arXiv 2306.00582 ↗First posted 1 Jun 2023cs.LG cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

We propose a modified density estimation problem that is highly effective for detecting anomalies in tabular data. Our approach assumes that the density function is relatively stable (with lower variance) around normal samples. We have verified this hypothesis empirically using a wide range of real-world data. Then, we present a variance-stabilized density estimation problem for maximizing the likelihood of the observed samples while minimizing the variance of the density around normal samples. To obtain a reliable anomaly detector, we introduce a spectral ensemble of autoregressive models for learning the variance-stabilized distribution. We have conducted an extensive benchmark with 52 datasets, demonstrating that our method leads to state-of-the-art results while alleviating the need for data-specific hyperparameter tuning. Finally, we have used an ablation study to demonstrate the importance of each of the proposed components, followed by a stability analysis evaluating the robustness of our model.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 28 of its 56 references that are in the map. First posted on arXiv 1 Jun 2023.

Builds on

28 of 56 references.

References · 28

  1. 2016Density estimation using Real NVPcited by 2,042
  2. 2020Energy-based Out-of-distribution Detectioncited by 1,299
  3. 2014NICE: Non-linear Independent Components Estimationcited by 1,058
  4. 2018Deep Anomaly Detection with Outlier Exposurecited by 942
  5. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  6. 2019Neural Spline Flowscited by 596
  7. 2017PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modificationscited by 417
  8. 2019Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Designcited by 262
  9. 2022ADBench: Anomaly Detection Benchmarkcited by 232
  10. 2020Classification-Based Anomaly Detection for General Datacited by 211
  11. 2019Deep Anomaly Detection with Deviation Networkscited by 153
  12. 2022ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functionscited by 153
16 more
  1. 2020COPOD: Copula-Based Outlier Detectioncited by 146
  2. 2020Deep Learning for Medical Anomaly Detection -- A Surveycited by 127
  3. 2021Neural Transformation Learning for Deep Anomaly Detection Beyond Imagescited by 99
  4. 2018Anomaly Detection using One-Class Neural Networkscited by 82
  5. 2020Anomaly Detection with Density Estimationcited by 72
  6. 2023On Diffusion Modeling for Anomaly Detectioncited by 72
  7. 2013RNADE: The real-valued neural autoregressive density-estimatorcited by 66
  8. 2022Latent Outlier Exposure for Anomaly Detection with Contaminated Datacited by 48
  9. 2020Perfect density models cannot guarantee anomaly detectioncited by 39
  10. 2019Noise Regularization for Conditional Density Estimationcited by 23
  11. 2019Nonparametric Density Estimation for High-Dimensional Data - Algorithms and Applicationscited by 18
  12. 2014Estimating the Accuracies of Multiple Classifiers Without Labeled Datacited by 17
  13. 2021Entropic Issues in Likelihood-Based OOD Detectioncited by 16
  14. 2022ButterflyFlow: Building Invertible Layers with Butterfly Matricescited by 12
  15. 2021Probabilistic Robust Autoencoders for Outlier Detectioncited by 7
  16. 2022Neural Inverse Transform Samplercited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 5; 2026: 1.

8 papers.

  1. 2025Representation Learning for Tabular Data: A Comprehensive Surveycited by 36
  2. 2023TabADM: Unsupervised Tabular Anomaly Detection with Diffusion Modelscited by 5
  3. 2024Unveiling Multiple Descents in Unsupervised Autoencoderscited by 5
  4. 2025Hybrid Autoencoders for Tabular Data: Leveraging Model-Based Augmentation in Low-Label Settingscited by 5
  5. 2025Multivariate Latent Recalibration for Conditional Normalizing Flowscited by 3
  6. 2025Detect and Correct: A Selective Noise Correction Method for Learning with Noisy Labelscited by 0
  7. 2025UniOD: A Universal Model for Outlier Detection across Diverse Domainscited by 0
  8. 2026When Identical Rows Disagree: From Benchmark Identifiability to Replication-Robust Anomaly Detectioncited by 0