BEGIN:VCALENDAR PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3- M3//EN VERSION:2.0 CALSCALE:GREGORIAN METHOD:PUBLISH BEGIN:VEVENT DTSTAMP:20260827T105636Z DTSTART:20260908T100000Z DTEND:20260908T110000Z SUMMARY:AI Fun with ELLIS Invited Speaker Series | Santiago Mazuelas Fran co UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}k5v-mta7hvk r-97edec DESCRIPTION:Our first invited speaker of 2026/27 will be Santiago Mazuela s Franco from Basque Center for Applied Mathematics. \n\nTitle: Beyond e mpirical risk minimization: performance guarantees\, distribution shifts \, and noise robustness\n\nAbstract: The empirical risk minimization (ER M) approach for supervised learning chooses prediction rules that fit tr aining samples and are “simple” (generalize). This approach has been the workhorse of machine learning methods and has enabled a myriad of appli cations. However\, ERM methods strongly rely on the specific training sa mples available and cannot easily address scenarios affected by distribu tion shifts or corrupted samples. Robust risk minimization (RRM) is an a lternative approach that does not aim to fit training examples and inste ad chooses prediction rules minimizing the maximum expected loss (risk). This talk presents a learning framework based on the generalized maximu m entropy principle that leads to minimax risk classifiers (MRCs). In pa rticular\, MRCs can minimize worst-case expected 0-1 loss while providin g performance guarantees\, and are strongly universally consistent using feature mappings given by characteristic kernels. In addition\, the met hods presented can provide techniques that are effective in practical si tuations that defy conventional assumptions\, such as scenarios affected by distribution shifts and corrupted samples.\n\nBio: Santiago Mazuelas received the Ph.D. in Mathematics and Ph.D. in Telecommunications Engin eering from the University of Valladolid\, Spain\, in 2009 and 2011\, re spectively. He is currently an Ikerbasque Research Professor at the Basq ue Center for Applied Mathematics (BCAM). Prior to joining BCAM\, he was a Staff Engineer at Qualcomm Corporate Research and Development from 20 14 to 2017. He previously worked from 2009 to 2014 as Postdoctoral Fello w and Associate in the Laboratory for Information and Decision Systems ( LIDS) at the Massachusetts Institute of Technology (MIT). Dr. Mazuelas i s currently Associate Editor-in-Chief for the IEEE Transactions on Mobil e Computing\, and serves as Area Chair at NeurIPS\, ICML\, AAAI\, and IC LR. He has received multiple awards including the IEEE Communications So ciety Fred W. Ellersick Prize in 2012\, the Early Achievement Award from the IEEE ComSoc in 2018\, and the SEIO-BBVA Foundation Best Contributio n in 2022 and 2025.\n\nIf you are unable to attend in person\, please re gister via the Ticketsource link provided and you will receive details f or joining via Teams.\n\n*Please note\, the date and room has now change d for this event since it was first posted on 26 Aug* STATUS:TENTATIVE TRANSP:TRANSPARENT CLASS:PUBLIC LOCATION:Alan Turing G.207\, Alan Turing Building\, Manchester END:VEVENT END:VCALENDAR