BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20251104T225727EST-57014fLknJ@132.216.98.100 DTSTAMP:20251105T035727Z DESCRIPTION:Camila P. E. de Souza\, PhD\n\nAssociate Professor\, Department of Statistical and Actuarial Sciences\n University of Western Ontario\n \n N OTE: Meet & Greet Prof Camila de Souza from 3-3:30pm in Room 1140\n\nWHEN: Wednesday\, November 5\, 2025\, from 3:30 to 4:30 p.m.\n WHERE: Hybrid | 2 001 ɬÀï·¬ College Avenue\, Rm 1140\; Zoom\n NOTE: Camila de Souza will be presenting in-person at SPGH\n  \n\nAbstract\n\nVariational Inference is a method for analytically approximating the posterior distribution in Bayesi an models\, offering a more computationally efficient alternative to Marko v Chain Monte Carlo (MCMC) sampling techniques. In this talk\, I will pres ent work from two recent publications\, co-authored with my students and c ollaborators. The first paper applies VI to functional data clustering\, w here the goal is to identify groups of curves without prior group membersh ip information. Using a B-spline regression mixture model with random inte rcepts\, we developed a novel variational Bayes (VB) algorithm for simulta neous clustering and smoothing of functional data. The second paper focuse s on survival data analysis\, proposing a VB algorithm for inferring the p arameters of the log-logistic accelerated failure time model by incorporat ing a piecewise approximation technique to address intractable calculation s and achieve Bayesian conjugacy. In both papers\, we conducted extensive simulation studies to assess the performance of the proposed VB algorithms \, comparing them with other methods\, including MCMC algorithms. Applicat ions to real data illustrate the practical use of the methodologies. The p roposed VB algorithms demonstrate excellent performance in clustering func tional data and analyzing survival data while significantly reducing compu tational costs compared to MCMC methods. The links to the papers are as fo llows: https://doi.org/10.1007/s11634-024-00590-w and https://doi.org/10.1 007/s11222-023-10365-6.\n \n Speaker Bio\n\nDr. Camila de Souza is an Associ ate Professor at the Department of Statistical and Actuarial Sciences at t he University of Western Ontario. Before joining Western\, Dr. de Souza wa s a postdoctoral fellow at the Shah Lab for Computational Cancer Biology a t the BC Cancer Agency Research Centre. She completed her PhD in Statistic s at the University of British Columbia (UBC). She is originally from Braz il\, where she received her Master’s and Bachelor’s Degrees in Statistics at the University of Campinas. Her research program consists of developing new statistical methods to analyze large and complex data structures aris ing from various areas in the Natural Sciences\, Health\, and Engineering. Dr. de Souza conducts research on techniques involving clustering\, hiera rchical mixture models\, mixed-effects models\, hidden Markov models\, non -parametric regression\, semi-parametric models\, the expectation-maximiza tion (EM) algorithm\, and Bayesian variational inference. Her website is h ttps://www.desouzacpe.com/\n  \n DTSTART:20251105T203000Z DTEND:20251105T213000Z SUMMARY:Advancing Functional Data Clustering and Survival Analysis with Var iational Inference URL:/spgh/channels/event/advancing-functional-data-clu stering-and-survival-analysis-variational-inference-368423 END:VEVENT END:VCALENDAR