PICS Colloquium: “Deep Anomaly Detection using Coincident Learning”

Abstract

Anomaly detection is a crucial task in the operation of complex systems such as industrial facilities, manufacturing plants, and large-scale science experiments. Failures in a sub-system can result in low yield, faulty products, or damage to components, making it essential to detect anomalies as quickly as possible. Despite the abundance of data available for complex systems, labeled anomalies are rare and expensive to obtain. To address this issue, we present a novel approach called CoAD that trains anomaly detection models on unlabeled data by leveraging the correlation between sub-systems and products. CoAD works by analyzing two data streams, s and q, which represent subsystem diagnostics and final product quality, respectively. We define an unsupervised metric, akin to the supervised classification F_beta statistic, to assess the performance of independent anomaly detection algorithms on s and q based on their coincidence rate. Our method is demonstrated in four cases, including a synthetic outlier data set, a synthetic imaging data set generated from MNIST, a metal milling data set, and a data set obtained from a particle accelerator. By using CoAD, we can detect anomalies in complex systems more effectively, even when labeled anomalies are scarce.

Bio

The research interests of Professor Darve span across several domains, including machine learning for engineering, surrogate and reduced order modeling, stochastic inversing, anomaly detection for engineering processes and manufacturing, numerical linear algebra, high-performance and parallel computing, and GPGPU.

Professor Darve received his Ph.D. in Applied Mathematics at the Jacques-Louis Lions Laboratory in the Pierre et Marie Curie University, Paris, France. His advisor was Prof. Olivier Pironneau, and his Ph.D. thesis was entitled “Fast Multipole Methods for Integral Equations in Acoustics and Electromagnetics.” He was previously a student at the Ecole Normale Supérieure, rue d’Ulm, Paris, in Mathematics and Computer Science.

Prof. Darve became a postdoctoral scholar with Profs. Moin and Pohorille at Stanford and NASA Ames in 1999 and joined the faculty at Stanford University in 2001. He is a member of the Institute for Computational and Mathematical Engineering.

Prof. Darve has received many awards, including the H. Julian Allen Award, NASA (2010), the Habilitation à Diriger des Recherches, France (2007), the Leslie Fox Prize in Numerical Analysis, IMA (2001), and the James H. Clark Faculty Scholar, Stanford University (2001).

Email jnespos@seas.upenn.edu for the Zoom link.