异常检测论文大列表：方法、应用、综述

导读

Github作者zhuyiche发布了一篇关于异常检测（Anomaly detection）的论文列表，包含异常检测领域的经典方法、深度学习方法、应用及综述。

作者 | zhuyiche

编译 | Xiaowen

https://github.com/zhuyiche/awesome-anomaly-detection

01

经典方法

• Isolation Forest - ICDM 2008.

https://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/icdm08b.pdf

• LOF: Identifying Density-Based Local Outliers - SIGMOD 2000.

http://www.dbs.ifi.lmu.de/Publikationen/Papers/LOF.pdf

• Extended Isolation Forest

http://matias-ck.com/files/papers/Extended_Isolation_Forest.pdf

• Support Vector Method for Novelty Detection - NIPS 2000

https://papers.nips.cc/paper/1723-support-vector-method-for-novelty-detection.pdf

One-Class Classification

• One-Class SVMs for Document Classification - JMLR 2001.

http://www.jmlr.org/papers/volume2/manevitz01a/manevitz01a.pdf

• Support Vector Data Description

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.100.1425&rep=rep1&type=pdf

• Can I Trust My One-Class Classification?

http://www.ipb.uni-bonn.de/pdfs/Mack2014Can.pdf

• Efficient Anomaly Detection via Matrix Sketching - NIPS 2018

https://arxiv.org/pdf/1804.03065.pdf

PCA-based

• robust deep and inductive anomaly detection - ECML PKDD 2017

https://arxiv.org/abs/1704.06743

• A loss framework for calibrated anomaly detection - NIPS 2018

https://papers.nips.cc/paper/7422-a-loss-framework-for-calibrated-anomaly-detection.pdf

Clustering

• A Practical Algorithm for Distributed Clustering and Outlier Detection - NIPS 2018

https://arxiv.org/pdf/1805.09495.pdf

Correlation

• Detecting Multiple Periods and Periodic Patterns in Event Time Sequences - CIKM 2017.

http://chaozhang.org/papers/cikm17a.pdf

Ranking

• ranking causal anomalies via temporal and dynamical analysis on vanishing correlations - KDD 2016.

https://www.kdd.org/kdd2016/papers/files/rfp0445-chengAemb.pdf

https://www.kdd.org/kdd2016/papers/files/rfp0445-chengAemb.pdf

02

深度学习方法

Generative Methods

• Variational Autoencoder based Anomaly Detection using Reconstruction Probability

http://dm.snu.ac.kr/static/docs/TR/SNUDM-TR-2015-03.pdf

Auto-encoder

• Learning sparse representation with variational auto-encoder for anomaly detection

https://ieeexplore.ieee.org/document/8386760/

• Anomaly Detection with Robust Deep Autoencoders - KDD 2017.

http://dl.acm.org/authorize?N33358

• DEEP AUTOENCODING GAUSSIAN MIXTURE MODEL FOR UNSUPERVISED ANOMALY DETECTION - ICLR 2018.

https://www.cs.ucsb.edu/~bzong/doc/iclr18-dagmm.pdf

• Generative Probabilistic Novelty Detection with Adversarial Autoencoders - NIPS 2018

https://papers.nips.cc/paper/7915-generative-probabilistic-novelty-detection-with-adversarial-autoencoders.pdf

Variational Auto-encoder

• Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach - ACML 2018

http://proceedings.mlr.press/v95/guo18a/guo18a.pdf

• A Multimodel Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencoder - IEEE Robotics and Automation Letters 2018.

https://arxiv.org/pdf/1711.00614.pdf

GAN based

• Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery - IPMI 2017.

https://arxiv.org/pdf/1703.05921.pdf

• Efficient-GAN-Based Anomaly Detection ICLR Workshop 2018.

https://github.com/houssamzenati/Efficient-GAN-Anomaly-

Detection

• Anomaly detection with generative adversarial networks - Reject by ICLR 2018, but was used as baseline method in recent published NIPS paper.

https://openreview.net/pdf?id=S1EfylZ0Z

Hypersphereical Learning

• Anomaly Detection in Dynamic Networks using Multi-view Time-Series Hypersphere Learning - CIKM 2017.

https://dl.acm.org/citation.cfm?id=3132964

• Deep into Hypersphere: Robust and Unsupervised Anomaly Discovery in Dynamic Networks - IJCAI 2018.

https://www.ijcai.org/proceedings/2018/0378.pdf

One-Class Classification

• High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning - Pattern Recognition 2018.

https://www.sciencedirect.com/science/article/abs/pii/S0031320316300267

• Optimal single-class classification strategies - NIPS 2007

https://papers.nips.cc/paper/2987-optimal-single-class-classification-strategies.pdf

• Deep One-Class Classification - ICML 2018.

http://proceedings.mlr.press/v80/ruff18a/ruff18a.pdf

Energy-based

• Deep structured energy based models for anomaly detection - ICML 2016

https://arxiv.org/pdf/1605.07717.pdf

Time series

• A Generalized Student-t Based Approach to Mixed-Type Anomaly Detection - AAAI 2013

http://www.nvc.cs.vt.edu/~ctlu/Publication/2013/AAAI-Lu-2013.pdf

• Stochastic Online Anomaly Analysis for Streaming Time Series - IJCAI 2017

https://www.ijcai.org/proceedings/2017/0445.pdf

• Long short term memory networks for anmomaly detection in time series

https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2015-56.pdf

• LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection - ICML 2016 Workshop.

https://arxiv.org/pdf/1607.00148.pdf

Interpretation

• Contextual Outlier Interpretation -IJCAI 2018

https://www.ijcai.org/proceedings/2018/0341.pdf

Evaulation Metrics

• Precision and Recall for Time Series - NIPS 2018

http://papers.nips.cc/paper/7462-precision-and-recall-for-time-series.pdf

Geometric transformation

• Deep Anomaly Detection Using Geometric Transformations - NIPS 2018

https://arxiv.org/pdf/1805.10917.pdf

FeedBack

• Incorporating Feedback into Tree-based Anomaly Detection - KDD 2017 Workshop on Interactive Data Exploration and Analytics.

https://github.com/ai/size-limit

• Feedback-Guided Anomaly Discovery via Online Optimization - KDD 2018.

http://web.engr.oregonstate.edu/~afern/papers/kdd18-siddiqui.pdf

03

异常检测应用

KPI

• Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications - WWW 2018.

https://arxiv.org/pdf/1802.03903

Log

• DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning - CCS 2017.

https://acmccs.github.io/papers/p1285-duA.pdf

• Mining Invariants from Logs for System Problem Detection - USENIX 2010

https://www.usenix.org/legacy/event/atc10/tech/slides/lou.pdf

https://www.usenix.org/legacy/event/atc10/tech/slides/lou.pdf

04

综述

Survey

• Anomaly detection in dynamic networks: a survey

https://onlinelibrary.wiley.com/doi/pdf/10.1002/wics.1347

• Anomaly Detection : A Survey

http://cucis.ece.northwestern.edu/projects/DMS/publications/AnomalyDetection.pdf

• A Survey of Recent Trends in One Class Classification

https://link.springer.com/chapter/10.1007/978-3-642-17080-5_21

• A survey on unsupervised outlier detection in high‐dimensional numerical data

https://onlinelibrary.wiley.com/doi/abs/10.1002/sam.11161

-END-

专 · 知

专知，专业可信的人工智能知识分发，让认知协作更快更好！欢迎登录www.zhuanzhi.ai，注册登录专知，获取更多AI知识资料！

欢迎微信扫一扫加入专知人工智能知识星球群，获取最新AI专业干货知识教程视频资料和与专家交流咨询！

请加专知小助手微信（扫一扫如下二维码添加），加入专知人工智能主题群，咨询技术商务合作~

专知《深度学习:算法到实战》课程全部完成！550+位同学在学习，现在报名，限时优惠！网易云课堂人工智能畅销榜首位！

点击“阅读原文”，了解报名专知《深度学习:算法到实战》课程