ADBench: Anomaly Detection Benchmark

Han, Songqiao; Hu, Xiyang; Huang, Hailiang; Jiang, Mingqi; Zhao, Yue

doi:10.48550/arxiv.2206.09426

Cited by 8 publications

(9 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We manipulated these components to generate anomalous samples that could be used to benchmark the performance of anomaly detection methods. This is an approach previously used by Han et al (2022) for evaluating anomaly detection methods on a series of 57 benchmark anomaly data sets across various practical application domains. This procedure ensures we can make systematic choices for the given anomaly detection task in order to find the optimal method.…”

Section: Gaussian Mixture Modeling Of the Orbit-color Spacementioning

confidence: 99%

The Weird and the Wonderful in Our Solar System: Searching for Serendipity in the Legacy Survey of Space and Time

Rogers,

Lintott,

Croft

et al. 2024

View full text Add to dashboard Cite

We present a novel method for anomaly detection in solar system object data in preparation for the Legacy Survey of Space and Time. We train a deep autoencoder for anomaly detection and use the learned latent space to search for other interesting objects. We demonstrate the efficacy of the autoencoder approach by finding interesting examples, such as interstellar objects, and show that by using the autoencoder, further examples of interesting classes can be found. We also investigate the limits of classic unsupervised approaches to anomaly detection through the generation of synthetic anomalies and evaluate the feasibility of using a supervised learning approach. Future work should consider expanding the feature space to increase the variety of anomalies that can be uncovered during the survey using an autoencoder.

show abstract

Section: Gaussian Mixture Modeling Of the Orbit-color Spacementioning

confidence: 99%

The Weird and the Wonderful in Our Solar System: Searching for Serendipity in the Legacy Survey of Space and Time

Rogers,

Lintott,

Croft

et al. 2024

View full text Add to dashboard Cite

show abstract

“…An effective semi-supervised anomaly detection algorithm should leverage label information to attain cutting-edge outcomes, all the while adhering to the unsupervised learning framework to detect previously unseen anomalies [14]. The benchmark in [15] shows that most semisupervised methods, with merely 1% of the data with labeled anomalies, can outperform the best unsupervised methods. The transfer learning framework proposed in this work is most closely related to the semi-supervised anomaly detection algorithms.…”

Section: A Anomaly Detectionmentioning

confidence: 99%

Transfer Learning for Anomaly Detection in Rotating Machinery using Data-driven Key Order Estimation

Liang,

Shui,

Gupta

et al. 2024

Preprint

View full text Add to dashboard Cite

Anomaly detection is an important task in industrial applications. However, designing an accurate anomaly detector can be very challenging in settings where anomalous labels are sparse or, in the worst case, missing in the training data. To mitigate this issue of a lack of anomalous labels in the domain of interest, existing approaches use transfer learning, leveraging information from anomalous samples in a closely related domain. Although previous studies have shown good results from applying transfer learning, they do not specifically address the issue of high false-positive rates, especially in industrial settings. High false-positive rates can arise from misleading information present in uninformative features. Inspired by this observation, the paper focuses on identifying key input features—termed as such due to their strong predictability in anomaly detection. A transfer learning approach is introduced that leverages the optimal fβ score for key feature estimation. This approach involves a weight vector to amplify key features and attenuate uninformative inputs during prediction. We demonstrate the capabilities of our proposed method through an industrial application: anomaly detection for rotating machinery. Based on our findings, anomaly detection algorithms that utilize data-driven features obtained through the proposed method outperform detectors based on features identified by domain experts. More importantly, our proposed framework can work with any downstream unsupervised anomaly detection algorithm, allowing us to freely choose the best algorithm for the anomaly detection task.

show abstract

“…There are many datasets publicly available for experimental analysis and anomaly detection. These datasets include both real-world benchmark datasets such as the collection of datasets presented in [16] 1 and synthetic data such as Sinusoidal and Blobs data sets. In this study, we have chosen to utilize a variety of real-world benchmark datasets (randomly Fig.…”

Section: A Data-setmentioning

confidence: 99%

Anomaly Detection using combination of Autoencoder and Isolation Forest

Almansoori¹,

Telek²

2023

1st Workshop on Intelligent Infocommunication Networks, Systems and Services

View full text Add to dashboard Cite

The process of identifying abnormal objects or patterns that deviate from the typical behavior in a dataset or other observations is known as Anomaly Detection. It is an essential technique in many fields, such as cyber security, finance, transportation, and fraud detection. This paper combines an autoencoder and an isolation forest algorithm to enhance anomaly detection where the individual methods might not perform well due to the specific context and the nature of the dataset. The autoencoder is a neural network trained to reconstruct the input data, while the isolation forest is a tree-based algorithm that can identify outliers in the data. By combining these two methods, the autoencoder can learn a compact representation of the data, and the isolation forest can then be applied to the reconstructed data to identify anomalies. This combination effectively enhances the anomaly detection process in high-dimensional data when compared to utilizing individual algorithms.

show abstract

ADBench: Anomaly Detection Benchmark

Cited by 8 publications

References 72 publications

The Weird and the Wonderful in Our Solar System: Searching for Serendipity in the Legacy Survey of Space and Time

The Weird and the Wonderful in Our Solar System: Searching for Serendipity in the Legacy Survey of Space and Time

Transfer Learning for Anomaly Detection in Rotating Machinery using Data-driven Key Order Estimation

Anomaly Detection using combination of Autoencoder and Isolation Forest

Contact Info

Product

Resources

About