Classification-Based Anomaly Detection for General Data

By Liron Bergman, Yedid Hoshen. International Conference on Learning Representations, 2020

425 citations recorded by the source corpus.

Anomaly detection, finding patterns that substantially deviate from those seen previously, is one of the fundamental problems of artificial intelligence. Recently, classification-based methods were shown to achieve superior results on this task. In this work, we present a unifying view and propose an open-set method to relax current generalization assumptions. Furthermore, we extend the applicability of transformation-based methods to non-image data using random affine transformations. Our method is shown to obtain state-of-the-art accuracy and is applicable to broad data types. The strong performance of our method is extensively validated on multiple datasets from different domains.

Subfields: Artificial Intelligence, Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/04513c7c0b3a63fde81a996dae064a28d453c17a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

Citations in the corpus

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances (2020, methodology): “1 GOAD∗ [2] WRN-10-4 77.”
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation (2020, background): “The seminal work by [10] was later extended by [15] and [1].”
Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation (2020, methodology): “[22] adopted an open-set classification method and generalized the method to include non-image data.”
CutPaste: Self-Supervised Learning for Anomaly Detection and Localization (2021, methodology): “One major family is by predicting geometric transformations [20, 24, 4], such as rotation, translation or flips.”
ADBench: Anomaly Detection Benchmark (2022, background): “Many unsupervised methods have been proposed in the last few decades [3, 15, 129, 150, 198], which can be roughly categorized into shallow and deep (neural network) methods.”
Multiresolution Knowledge Distillation for Anomaly Detection (2020, methodology): “Training using Ltotal, unlike many other methods [18, 6], continues to fully converge, which is the only accessible criterion to measure when to stop training epochs.”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “pseudolabeling [152], [153], [155], [335], or integrating some manifold assumption [333].”
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows (2021, methodology): “…networks to obtain the features of normal images and model the distribution with some statistical methods, then detect the abnormal samples that have different distributions (Bergman and Hoshen 2020; Rippel, Mertens, and Merhof 2021; Yi and Yoon 2020; Cohen and Hoshen 2020; Defard et al. 2020).”
CMG: A Class-Mixed Generation Approach to Out-of-Distribution Detection (2022, neutral): “So far, many approaches have been proposed to solve this problem, from distance-based meth-ods [2, 3, 12, 20], to generative models [60, 38, 41, 36] and self-supervised learning methods [4, 11, 17, 19].”
Generalized Out-of-Distribution Detection: A Survey (2021, positive): “Self-supervised learning methods use pretext tasks such as contrastive learning (Tack et al., 2020), image transformation prediction (Bergman & Hoshen, 2020; Golan & El-Yaniv, 2018), and future frame prediction (Georgescu et al., 2021), where anomalies are more likely to make mistakes on the…”
Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detection (2020, result): “Such pretraining is consistent with recent anomaly detection literature, which suggests that training on the “normal” class (in this case real videos) for a suitable task promotes learning of features that are useful for detecting “anomalous” samples (in this case fake videos) [23, 7].”
Sub-Image Anomaly Detection with Deep Pyramid Correspondences (2020, methodology): “[4] illustrates that self-supervised features underperform ImageNet-trained ResNet features for the purposes of anomaly detection.”