BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20250707T150033EDT-8855psL4O3@132.216.98.100 DTSTAMP:20250707T190033Z DESCRIPTION:Title: Probabilistic Approaches to Machine Learning on Tensor D ata\n\n\n Abstract:\n\n\nIn contemporary scientific research\, it is often of great interest to predict a categorical response based on a high-dimens ional tensor (i.e. multi-dimensional array). Motivated by applications in science and engineering\, we propose two probabilistic methods for machine learning on tensor data in the supervised and the unsupervised context\, respectively. For supervised problems\, we develop a comprehensive discrim inant analysis model\, called the CATCH model. The CATCH model integrates the information from the tensor and additional covariates to predict the c ategorical outcome with high accuracy. We further consider unsupervised pr oblems\, where no categorical response is available even on the training d ata. A doubly-enhanced EM (DEEM) algorithm is proposed for model-based ten sor clustering\, in which both the E-step and the M-step are carefully tai lored for tensor data. CATCH and DEEM are developed under explicit statist ical models with clear interpretations. They aggressively take advantage o f the tensor structure and sparsity to tackle the new computational and st atistical challenges arising from the intimidating tensor dimensions. Effi cient algorithms are developed to solve the related optimization problems. Under mild conditions\, CATCH and DEEM are shown to be consistent even wh en the dimension of each mode grows at an exponential rate of the sample s ize. Numerical studies also strongly support the application of CATCH and DEEM.\n\n\n Speaker\n\n\nQing Mai is an associate professor in the Departme nt of Statistics at Florida State University.\n\nZoom Link\n\nMeeting ID: 924 5390 4989\n\nPasscode: 690084\n\n \n DTSTART:20201127T203000Z DTEND:20201127T213000Z SUMMARY:Qing Mai (Florida State University) URL:/mathstat/channels/event/qing-mai-florida-state-un iversity-326442 END:VEVENT END:VCALENDAR