mlsanityData to 14 Oct 2026

Bridging Unsupervised and Semi-Supervised Anomaly Detection: A Theoretically-Grounded and Practical Framework with Synthetic Anomalies

arXiv 2506.13955 ↗First posted 16 Jun 2025stat.ML cs.CR cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Anomaly detection (AD) is a critical task across domains such as cybersecurity and healthcare. In the unsupervised setting, an effective and theoretically-grounded principle is to train classifiers to distinguish normal data from (synthetic) anomalies. We extend this principle to semi-supervised AD, where training data also include a limited labeled subset of anomalies possibly present in test time. We propose a theoretically-grounded and empirically effective framework for semi-supervised AD that combines known and synthetic anomalies during training. To analyze semi-supervised AD, we introduce the first mathematical formulation of semi-supervised AD, which generalizes unsupervised AD. Here, we show that synthetic anomalies enable (i) better anomaly modeling in low-density regions and (ii) optimal convergence guarantees for neural network classifiers -- the first theoretical result for semi-supervised AD. We empirically validate our framework on five diverse benchmarks, observing consistent performance gains. These improvements also extend beyond our theoretical framework to other classification-based AD methods, validating the generalizability of the synthetic anomaly principle in AD.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 53 references that are in the map. First posted on arXiv 16 Jun 2025.

Builds on

23 of 53 references.

References · 23

  1. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  2. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  3. 2018Deep Anomaly Detection with Outlier Exposurecited by 942
  4. 2016Error bounds for approximations with deep ReLU networkscited by 574
  5. 2017Nonparametric regression using deep neural networks with ReLU activation functioncited by 433
  6. 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 425
  7. 2019Deep Semi-Supervised Anomaly Detectioncited by 291
  8. 2022ADBench: Anomaly Detection Benchmarkcited by 232
  9. 2019Deep Anomaly Detection with Deviation Networkscited by 153
  10. 2020DROCC: Deep Robust One-Class Classificationcited by 121
  11. 2022Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLPcited by 102
  12. 2021Neural Transformation Learning for Deep Anomaly Detection Beyond Imagescited by 99
11 more
  1. 2021Optimal Approximation Rate of ReLU Networks in terms of Width and Depthcited by 96
  2. 2021Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detectioncited by 42
  3. 2020Interpretable, Multidimensional, Multimodal Anomaly Detection with Negative Sampling for Detection of Device Failurecited by 27
  4. 2024AD-LLM: Benchmarking Large Language Models for Anomaly Detectioncited by 26
  5. 2022Perturbation Learning Based Anomaly Detectioncited by 24
  6. 2023NNG-Mix: Improving Semi-supervised Anomaly Detection with Pseudo-anomaly Generationcited by 14
  7. 2023Training shallow ReLU networks on noisy data using hinge loss: when do we overfit and is it benign?cited by 7
  8. 2019Bias of Homotopic Gradient Descent for the Hinge Losscited by 3
  9. 2023Revisiting Non-separable Binary Classification and its Applications in Anomaly Detectioncited by 3
  10. 2017Learning rates for classification with Gaussian kernelscited by 1
  11. 2024Learning to Detect Cyber Attacks: Neural Anomaly Detection for Cybersecurity with Theoretical Insightscited by 1

Built on by

Not cited yet.