paper-with-me

Papers

Is AUC the best measure for practical comparison of anomaly detectors?

2023-05-08 · Vít Škvára, Tomáš Pevný, Václav Šmídl

The area under receiver operating characteristics (AUC) is the standard measure for comparison of anomaly detectors. Its advantage is in providing a scalar number that allows a natural ordering and is independent on a threshold, which allows to postpone the choice. In this work, we question whether AUC is a good metric for anomaly detection, or if it gives a false sense of comfort, due to relying on assumptions which are unlikely to hold in practice. Our investigation shows that variations of AUC emphasizing accuracy at low false positive rate seem to be better correlated with the needs of practitioners, but also that we can compare anomaly detectors only in the case when we have representative examples of anomalous samples. This last result is disturbing, as it suggests that in many cases, we should do active or few-show learning instead of pure anomaly detection.

📄 PDF Abstract BibTeX arXiv:2305.04754

Code (1)

vitskvara/admetricevaluation.jl 공식 구현

Tasks

Anomaly Detection

Similar Papers 제목 키워드 기반

Evaluating Anomaly Detectors for Simulated Highly Imbalanced Industrial Classification Problems

2025-12-07 · Lesley Wheat, Martin v. Mohrenschildt, Saeid Habibi arxiv

Machine learning offers potential solutions to current issues in industrial systems in areas such as quality control and predictive maintenance, but also faces unique barriers in industrial applications. An ongoing chall…

Anomaly Detection

Training-Free Logical and Structural Anomaly Detection via Calibrated Fusion

2026-09-04 · Changyi Li, Miao Yu, Kai Dong, Yu Xiao arxiv

Industrial anomaly detection must handle two distinct defect families: structural anomalies, which manifest as local texture corruptions, and logical anomalies, which violate global rules on object count, composition, or…

Anomaly Detection

Comparison of Anomaly Detectors: Context Matters

2020-12-11 · Vít Škvára, Jan Franců, Matěj Zorek, Tomáš Pevný 외

Deep generative models are challenging the classical methods in the field of anomaly detection nowadays. Every new method provides evidence of outperforming its predecessors, often with contradictory results. The objecti…

Anomaly DetectionModel Selection

Experience Report: Deep Learning-based System Log Analysis for Anomaly Detection

2021-07-13 · Zhuangbin Chen, Jinyang Liu, Wenwei Gu, Yuxin Su 외

Logs have been an imperative resource to ensure the reliability and continuity of many software systems, especially large-scale distributed systems. They faithfully record runtime information to facilitate system trouble…

Anomaly Detection

ECoLAD: Selecting Anomaly Detectors for Automotive Deployment via Compute-Reduction Evaluation

2026-03-11 · Kadir-Kaan Özer, René Ebeling, Markus Enzweiler arxiv

Automotive anomaly detectors are often selected from accuracy only benchmarks on workstation class hardware, whereas in-vehicle monitoring requires predictable scoring latency under limited CPU parallelism. This mismatch…

Anomaly Detection