BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2// METHOD:PUBLISH BEGIN:VTIMEZONE TZID:America/New_York BEGIN:STANDARD DTSTART:20231105T020000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:EST END:STANDARD BEGIN:DAYLIGHT DTSTART:20230312T020000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RDATE:20240310T020000 TZNAME:EDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:calendar.2608.events_uoft_date.0@www.statistics.utoronto.ca CREATED:20230830T150958Z DESCRIPTION:\nWhen and Where: \nMonday, September 25, 2023 3:30 pm to 4:3 0 pm \n 9014 \n Ontario Power Generation \n 9-700 University ave, Toronto , ON M5G 1Z5 \n\nDescription: \nJoin us at the Statistical Sciences Appli ed Research and Education Seminar (ARES) with Stephen PortilloAssistant Pr ofessor of Physics Concordia University of EdmontonFree Hybrid (In-person/ Online) Event | Registration Required Talk TitleFrom Pixels to Parameters AbstractMuch of astronomy uses pixelized data, but the size and complexit y of these data often strain the capability of existing data analysis tech niques. I will present algorithms built on advances in statistics and mach ine learning that allow more science to be done with the same pixels. Digi tal tracking searches for Kuiper belt objects (KBOs) involve series of ima ges where the KBO is undetectable in each image, but detectable in the se ries. By forward modelling the position of the KBO in each image in a “joi nt-fit”, the KBOs’ trajectories can be measured precisely enough to const rain their dynamics. Probabilistic cataloguing (PCat) is a reversible jump Markov chain Monte Carlo method that creates model images for an unknown number of sources. In extremely crowded fields with stars every 10 pixels\ , PCat finds stars four times fainter than DAOPHOT, a commonly used pipel ine. Finally, I will discuss a variational autoencoder, a type of deep g enerative model, that I have trained on galaxy spectra from the Sloan Dig ital Sky Survey. This autoencoder learns a six dimensional latent space th at naturally separates different classes of galaxy and captures variation in spectral line widths and ratios. Speaker ProfileDr. Portillo is an Assi stant Professor of Physics at Concordia University of Edmonton. He obtaine d his PhD in Astronomy and Astrophysics from Harvard University and was a DIRAC Postdoctoral Fellow at the University of Washington. His research ap plies advances in statistics and machine learning to enable more science t o be done with astronomical data sets. The techniques he uses in his resea rch include Bayesian inference, reversible jump Markov chain Monte Carlo\ , and variational autoencoders. \n\nContact Information: \n Esther Berzunz a esther.berzunza@utoronto.ca 4166897271 CANSSI Ontario \n9-700 University ave, Toronto, ON M5G 1Z5 \n\nCategories \n Data Science ARESSeminar Ser ies \n\nAudiences \n FacultyGraduate Students DTSTART;TZID=America/New_York:20230925T153000 DTEND;TZID=America/New_York:20230925T163000 LAST-MODIFIED:20250401T204345Z LOCATION:9-700 University ave, Toronto, ON M5G 1Z5 SUMMARY:Statistical Sciences ARES: Stephen Portillo URL;TYPE=URI:https://www.statistics.utoronto.ca/events/statistical-sciences -ares-stephen-portillo-0 END:VEVENT END:VCALENDAR