Generative Adversarial Active Learning for Unsupervised Outlier Detection

By Yezheng Liu, Zhe Li, Chong Zhou, Yuanchun Jiang, Jianshan Sun, M. Wang, Xiangnan He. IEEE Transactions on Knowledge and Data Engineering, 2018

355 citations recorded by the source corpus.

Outlier detection is an important topic in machine learning and has been used in a wide range of applications. In this paper, we approach outlier detection as a binary-classification issue by sampling potential outliers from a uniform reference distribution. However, due to the sparsity of data in high-dimensional space, a limited number of potential outliers may fail to provide sufficient information to assist the classifier in describing a boundary that can separate outliers from normal data effectively. To address this, we propose a novel Single-Objective Generative Adversarial Active Learning (SO-GAAL) method for outlier detection, which can directly generate informative potential outliers based on the mini-max game between a generator and a discriminator. Moreover, to prevent the generator from falling into the mode collapsing problem, the stop node of training should be determined when SO-GAAL is able to provide sufficient information. But without any prior information, it is extremely difficult for SO-GAAL. Therefore, we expand the network structure of SO-GAAL from a single generator to multiple generators with different objectives (MO-GAAL), which can generate a reasonable reference distribution for the whole dataset. We empirically compare the proposed approach with several state-of-the-art outlier detection methods on both synthetic and real-world datasets. The results show that MO-GAAL outperforms its competitors in the majority of cases, especially for datasets with various cluster types or high irrelevant variable ratio. The experiment codes are available at: https://github.com/leibinghe/GAAL-based-outlier-detection.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/17eeaa7bfbadbcd92d5bacdda2e2f5760454eed9

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

Citations in the corpus

Deep Learning for Anomaly Detection (2020, background): “nstances. For example, uniformly distributed instances can be generated to enforce the normal instances to be distributed uniformly across the latent space [121]; an ensemble of generators is used in [91], with each generator synthesizing boundary instances for one specific cluster of normal instanc”
OpenGAN: Open-Set Recognition via Open Data Generation (2021, critical): “While this has been used as a baseline which does not work well in the literature [47, 44, 39, 56, 30], to the best of our knowledge, it is the first time that GAN-discriminator outperforms prior art on various benchmarks, thanks to the model selection via open validation (Section 3.”
PyOD: A Python Toolbox for Scalable Outlier Detection (2019, background): “, 2018) Neural Net No No MO-GAAL (Liu et al., 2018) Neural Net No No XGBOD (Zhao and Hryniewicki, 2018b) Ensembling No Yes LSCP (Zhao et al.”
COPOD: Copula-Based Outlier Detection (2020, methodology): “Numerous algorithms have been proposed over the last decades, which includes covariance based models like Outlier Detection with Minimum Covariance Determinant (MCD) [3], proximity based models like k-Nearest Neighbors Detector (kNN) [4], linear models like One-class SVM detector [5], ensemble base ”
ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions (2022, background): “Numerous unsupervised OD algorithms have been proposed over the years (e.g., [15], [16], [17], [18], [19], [20]; we provide a more detailed overview in Section 2).”
A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications (2020, neutral): “There are many papers related to GANs [114], [115], [116], [117], [118], [119], [120], [121], [122], [123], [124], [125], [126],”
Progress in Outlier Detection Techniques: A Survey (2019, background): “Active learning for outlier detection has recently been embraced in different research domain [195]–[199].”