Anomaly Detection
Anomaly detection focuses on identifying unusual patterns or deviations from expected behavior within data, aiming to improve system reliability and safety across diverse applications. Current research emphasizes unsupervised and self-supervised learning approaches, employing architectures like autoencoders, transformers, and graph neural networks, often incorporating techniques such as Bayesian inference and metric learning to enhance robustness and interpretability. The field's significance stems from its broad applicability, ranging from fraud detection and medical diagnosis to industrial process monitoring and network security, with ongoing efforts to develop more efficient, accurate, and explainable methods.
Papers
Challenges and Solutions to Build a Data Pipeline to Identify Anomalies in Enterprise System Performance
Xiaobo Huang, Amitabha Banerjee, Chien-Chia Chen, Chengzhi Huang, Tzu Yi Chuang, Abhishek Srivastava, Razvan Cheveresan
Why Are You Weird? Infusing Interpretability in Isolation Forest for Anomaly Detection
Nirmal Sobha Kartha, Clément Gautrais, Vincent Vercruyssen
Fast and scalable neuroevolution deep learning architecture search for multivariate anomaly detection
M. Pietroń, D. Żurek, K. Faber, R. Corizzo
Multimedia Datasets for Anomaly Detection: A Review
Pratibha Kumari, Anterpreet Kaur Bedi, Mukesh Saini
LUNAR: Unifying Local Outlier Detection Methods via Graph Neural Networks
Adam Goodge, Bryan Hooi, See Kiong Ng, Wee Siong Ng