# Perturbation Learning Based Anomaly Detection

arXiv: https://arxiv.org/abs/2206.02704 · HTML: https://mlsanity.com/paper/2206.02704/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 48 references that are in the map.
- First posted on arXiv 6 Jun 2022.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Deep Clustering for Unsupervised Learning of Visual Features](https://mlsanity.com/paper/1807.05520/) (2018)
- [A Unifying Review of Deep and Shallow Anomaly Detection](https://mlsanity.com/paper/2009.11732/) (2020)
- [Deep Semi-Supervised Anomaly Detection](https://mlsanity.com/paper/1906.02694/) (2019)
- [Deep Anomaly Detection Using Geometric Transformations](https://mlsanity.com/paper/1805.10917/) (2018)
- [LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection](https://mlsanity.com/paper/1607.00148/) (2016)
- [Classification-Based Anomaly Detection for General Data](https://mlsanity.com/paper/2005.02359/) (2020)
- [OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representations](https://mlsanity.com/paper/1903.08550/) (2019)
- [Latent Space Autoregression for Novelty Detection](https://mlsanity.com/paper/1807.01653/) (2018)
- [DROCC: Deep Robust One-Class Classification](https://mlsanity.com/paper/2002.12718/) (2020)
- [Generative Probabilistic Novelty Detection with Adversarial Autoencoders](https://mlsanity.com/paper/1807.02588/) (2018)
- [Neural Transformation Learning for Deep Anomaly Detection Beyond Images](https://mlsanity.com/paper/2103.16440/) (2021)
- [A Survey of Single-Scene Video Anomaly Detection](https://mlsanity.com/paper/2004.05993/) (2020)
- [Learning Deep Features for One-Class Classification](https://mlsanity.com/paper/1801.05365/) (2018)
- [Deep One-Class Classification via Interpolated Gaussian Descriptor](https://mlsanity.com/paper/2101.10043/) (2021)
- [A General Framework For Detecting Anomalous Inputs to DNN Classifiers](https://mlsanity.com/paper/2007.15147/) (2020)
- [Deep Anomaly Detection for Generalized Face Anti-Spoofing](https://mlsanity.com/paper/1904.08241/) (2019)

## Built on by (most cited first, 24 of 24)

- [ARC: A Generalist Graph Anomaly Detector with In-Context Learning](https://mlsanity.com/paper/2405.16771/) (2024)
- [Calibrated One-class Classification for Unsupervised Time Series Anomaly Detection](https://mlsanity.com/paper/2207.12201/) (2022)
- [GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features](https://mlsanity.com/paper/2407.12427/) (2024)
- [Generative Semi-supervised Graph Anomaly Detection](https://mlsanity.com/paper/2402.11887/) (2024)
- [NNG-Mix: Improving Semi-supervised Anomaly Detection with Pseudo-anomaly Generation](https://mlsanity.com/paper/2311.11961/) (2023)
- [Deep Orthogonal Hypersphere Compression for Anomaly Detection](https://mlsanity.com/paper/2302.06430/) (2023)
- [PseudoBound: Limiting the anomaly reconstruction capability of one-class classifiers using pseudo anomalies](https://mlsanity.com/paper/2303.10704/) (2023)
- [Anomaly Detection and Generation with Diffusion Models: A Survey](https://mlsanity.com/paper/2506.09368/) (2025)
- [Adaptive Multi-prompt Contrastive Network for Few-shot Out-of-distribution Detection](https://mlsanity.com/paper/2506.17633/) (2025)
- [Scale-Aware Contrastive Reverse Distillation for Unsupervised Medical Anomaly Detection](https://mlsanity.com/paper/2503.13828/) (2025)
- [Subject Information Extraction for Novelty Detection with Domain Shifts](https://mlsanity.com/paper/2504.21247/) (2025)
- [Unilaterally Aggregated Contrastive Learning with Hierarchical Augmentation for Anomaly Detection](https://mlsanity.com/paper/2308.10155/) (2023)
- [Restricted Generative Projection for One-Class Classification and Anomaly Detection](https://mlsanity.com/paper/2307.04097/) (2023)
- [Removing Geometric Bias in One-Class Anomaly Detection with Adaptive Feature Perturbation](https://mlsanity.com/paper/2503.05520/) (2025)
- [Interpretable Maximum Margin Deep Anomaly Detection](https://mlsanity.com/paper/2603.07073/) (2026)
- [Multi-Relation Graph-Kernel Strengthen Network for Graph-Level Clustering](https://mlsanity.com/paper/2504.01605/) (2025)
- [Fairness-aware Anomaly Detection via Fair Projection](https://mlsanity.com/paper/2505.11132/) (2025)
- [Text-ADBench: Text Anomaly Detection Benchmark Based on LLM Embeddings](https://mlsanity.com/paper/2507.12295/) (2025)
- [AnoRand: A Semi Supervised Deep Learning Anomaly Detection Method by Random Labeling](https://mlsanity.com/paper/2305.18389/) (2023)
- [Dual Boost-Driven Graph-Level Clustering Network](https://mlsanity.com/paper/2504.05670/) (2025)
- [Bridging Unsupervised and Semi-Supervised Anomaly Detection: A Theoretically-Grounded and Practical Framework with Synthetic Anomalies](https://mlsanity.com/paper/2506.13955/) (2025)
- [UniOD: A Universal Model for Outlier Detection across Diverse Domains](https://mlsanity.com/paper/2507.06624/) (2025)
- [Make Some Noise: Unsupervised Remote Sensing Change Detection Using Latent Space Perturbations](https://mlsanity.com/paper/2602.19881/) (2026)
- [Enhancing Tabular Anomaly Detection via Pseudo-Label-Guided Generation](https://mlsanity.com/paper/2604.18266/) (2026)

## Abstract (arXiv)

This paper presents a simple yet effective method for anomaly detection. The main idea is to learn small perturbations to perturb normal data and learn a classifier to classify the normal data and the perturbed data into two different classes. The perturbator and classifier are jointly learned using deep neural networks. Importantly, the perturbations should be as small as possible but the classifier is still able to recognize the perturbed data from unperturbed data. Therefore, the perturbed data are regarded as abnormal data and the classifier provides a decision boundary between the normal data and abnormal data, although the training data do not include any abnormal data. Compared with the state-of-the-art of anomaly detection, our method does not require any assumption about the shape (e.g. hypersphere) of the decision boundary and has fewer hyper-parameters to determine. Empirical studies on benchmark datasets verify the effectiveness and superiority of our method.