R2-AD2: Detecting Anomalies by Analysing the Raw Gradient
Not on the map. Why?
Abstract (arXiv)
Neural networks follow a gradient-based learning scheme, adapting their mapping parameters by back-propagating the output loss. Samples unlike the ones seen during training cause a different gradient distribution. Based on this intuition, we design a novel semi-supervised anomaly detection method called R2-AD2. By analysing the temporal distribution of the gradient over multiple training steps, we reliably detect point anomalies in strict semi-supervised settings. Instead of domain dependent features, we input the raw gradient caused by the sample under test to an end-to-end recurrent neural network architecture. R2-AD2 works in a purely data-driven way, thus is readily applicable in a variety of important use cases of anomaly detection.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 31 of its 46 references that are in the map. First posted on arXiv 21 Jun 2022.
Builds on
31 of 46 references.
References · 31
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
- 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
- 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
- 2018Deep Anomaly Detection with Outlier Exposurecited by 939
- 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 423
- 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
- 2019Deep Learning for Anomaly Detection: A Surveycited by 357
- 2021On the Importance of Gradients for Detecting Distributional Shifts in the Wildcited by 338
- 2019Deep Semi-Supervised Anomaly Detectioncited by 290
- 2018Deep Anomaly Detection Using Geometric Transformationscited by 269
- 2020Classification-Based Anomaly Detection for General Datacited by 210
- 2019MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networkscited by 176
19 more
- 2021A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challengescited by 154
- 2019Deep Anomaly Detection with Deviation Networkscited by 152
- 2020Learning and Evaluating Representations for Deep One-class Classificationcited by 133
- 2020DROCC: Deep Robust One-Class Classificationcited by 121
- 2018Anomaly Detection using One-Class Neural Networkscited by 82
- 2018Learning Representations of Ultrahigh-dimensional Data for Random Distance-based Outlier Detectioncited by 79
- 2020Backpropagated Gradient Representations for Anomaly Detectioncited by 54
- 2019Robust Anomaly Detection in Images using Adversarial Autoencoderscited by 38
- 2018Anomaly Detection using Autoencoders in High Performance Computing Systemscited by 25
- 2020GraN: An Efficient Gradient-Norm Based Detector for Adversarial and Misclassified Examplescited by 18
- 2019Autoencoding Binary Classifiers for Supervised Anomaly Detectioncited by 17
- 2020Novelty Detection Through Model-Based Characterization of Neural Networkscited by 17
- 2019Anomaly Detection with Adversarial Dual Autoencoderscited by 16
- 2022Gradient-Based Adversarial and Out-of-Distribution Detectioncited by 14
- 2021Gradient-based Novelty Detection Boosted by Self-supervised Binary Classificationcited by 10
- 2020ESAD: End-to-end Deep Semi-supervised Anomaly Detectioncited by 8
- 2021Understanding the Effect of Bias in Deep Anomaly Detectioncited by 7
- 2022Open-Set Recognition with Gradient-Based Representationscited by 7
- 2018Gradient Similarity: An Explainable Approach to Detect Adversarial Attacks against Deep Learningcited by 6
Built on by
Citing papers per year: 2023: 1; 2024: 1.
2 papers.