JSM2025
Back to the program
Contributed Papers

Anomaly Detection and Robust Modeling

Thu, Aug 7, 10:30 AM - 12:20 PM Room CC-208A Music City Center
Jing OuyangChair
Section on Statistical Learning and Data Science

7 Presentations

Zooming in on Anomalies: An Approach to Detecting Cellwise Outliers
10:35 AM - 10:50 AM
Co-authors: Jackson Passey, William Christensen (Brigham Young University), Jackson Passey
cellwise outliersanomaly detectionexception miningmultivariate dataMahalanobis' distance
Distributionally Robust Posterior Sampling - A Variational Bayes Approach
10:50 AM - 11:05 AM
Co-authors: Bohan Wu (Columbia University), Bennett Zhu, Bennett Zhu, David Blei (Columbia University-Data Science Institute)
Variational BayesDistributionally Robust InferenceWasserstein-Fisher-Rao Gradient flowmixed global-local contaminationadversarial contamination
Estimating Counts Through an Average Rounded to an Integer and its Theoretical & Practical Effects
11:05 AM - 11:20 AM
Co-authors: Roberto Rivera (University of Puerto Rico-Mayaguez), Axel Cortes-Cubero (Protocol Labs), Roberto Rivera (University of Puerto Rico-Mayaguez), Israel Almodovar-Rivera (University of Puerto Rico At Mayaguez), Wolfgang Rolke (University of Puerto Rico RUM)
rounding errorbinningSheppard’s correctiondiscrete Fourier transformexcess deathsprobability generating function
Hyper-Parameter Selection for Robust & Non-Convex Estimators via Information Sharing
11:20 AM - 11:35 AM
Co-authors: David Kepplinger (George Mason University), Siqi Wei (George Mason University), David Kepplinger (George Mason University)
Robust regressionHyper-parameter tuningCross-validationNon-convexity estimatorPenalized regression
Robust Multitask Feature Learning with Adaptive Huber Regressions
11:35 AM - 11:50 AM
Co-authors: Yuan Zhong, Yuan Zhong, Wei Xu (University of Toronto), Xin Gao (York University)
Adaptive Huber LossMultitask Feature LearningRobust M-estimationHeavy-tailed Data Integration
The Theory and Practice of Anomaly Detection using Neural Networks: Benefit of Synthetic Data
11:50 AM - 12:05 PM
Co-authors: Tian-Yi Zhou, Matthew Lau (Georgia Institute of Technology), Tian-Yi Zhou, Xiaoming Huo (Georgia Institute of Technology, School of Industrial & Systems Engineering), Jizhou Chen (Georgia Institute of Technology), Xiangchi Yuan (Georgia Institute of Technology), Wenke Lee (Georgia Institute of Technology)
Deep Learning TheoryAnomaly DetectionCybersecurityLLM SafetyClassification
Anomaly Detection in Neural Networks via One-Class Support Vector Methods
12:05 PM - 12:20 PM
Co-authors: Edgard M. Maboudou-Tchao (University of Central Florida), Poorna Sandamini Senaratne (University of Central Florida), Poorna Sandamini Senaratne (University of Central Florida)
Embedding LayerANNOne-class ClassificationLS-SVDDSVDD
↑