*Sam Staton giving this year's LMS/BCS-FACS Evening Seminar (online) -- registration open until this Wednesday at 5PM, UTC@ 2022-11-14 12:49 Andrei Popescu0 siblings, 0 replies; 2+ messages in thread From: Andrei Popescu @ 2022-11-14 12:49 UTC (permalink / raw) To: caml-list, categories, cl-isabelle-users, concurrency EVENT: Annual LMS/BCS-FACS Evening Seminar SPEAKER: Sam Staton, University of Oxford TITLE: Programming-based foundations for statistics DATE: Thursday, 17 November 2022, Starting time: 18:00 UTC VENUE: Online via Zoom EVENT PAGE: https://www.lms.ac.uk/events/lectures/lms-bcs-facs-evening-seminars REGISTRATION LINK: https://www.lms.ac.uk/civicrm/event/register?id=88&reset=1 Prior registration is required for attendance -- the registration site will close on Wednesday, 16 November, at 17:00 SYNOPSIS Probabilistic programming is a popular tool for statistics and machine learning. The idea is to describe a statistical model as a program with random choices. The program might be a simulation of a system, such as a physics model, a model of viral spread, or a model of electoral behaviour. We can now carry out statistical inference over the system, for example, by running a Monte Carlo simulation – running the simulation 100,000’s of times. As I will discuss in this talk, the idea of treating statistical models as computer programs also has a foundational appeal. If we can understand statistical models as programs, then the foundations of probability and statistics can be discussed in terms of program semantics. There is a chance of new foundational perspectives on statistics, in terms of programming languages and their formal methods. As I will explain, this programming-based foundation for statistics is attractive because there are some intuitively simple scenarios, such as inference over function spaces, which have an easy programming implementation, but for which the traditional mathematical interpretation is complicated. SPEAKER BIOGRAPHY Sam Staton is a Professor of Computer Science and Royal Society University Research Fellow at the University of Oxford. There he currently runs an ERC grant "Better Languages for Statistics". Before arriving in Oxford in 2015, Sam spent time in Nijmegen, Paris, and Cambridge. His PhD was in Cambridge with Marcelo Fiore (2007). Sam's main research is in programming language theory, but he is also interested in logic and category theory. He has recent contributions in probabilistic programming languages, and quantum computing and programming languages. Webpage: https://www.cs.ox.ac.uk/people/samuel.staton/main.html ^ permalink raw reply [flat|nested] 2+ messages in thread

*Sam Staton giving this year's LMS/BCS-FACS Evening Seminar (online) -- registration open until this Wednesday at 5PM, UTC@ 2022-11-14 12:49 Andrei Popescu0 siblings, 0 replies; 2+ messages in thread From: Andrei Popescu @ 2022-11-14 12:49 UTC (permalink / raw) To: categories EVENT: Annual LMS/BCS-FACS Evening Seminar SPEAKER: Sam Staton, University of Oxford TITLE: Programming-based foundations for statistics DATE: Thursday, 17 November 2022, Starting time: 18:00 UTC VENUE: Online via Zoom EVENT PAGE: https://www.lms.ac.uk/events/lectures/lms-bcs-facs-evening-seminars REGISTRATION LINK: https://www.lms.ac.uk/civicrm/event/register?id=88&reset=1 Prior registration is required for attendance -- the registration site will close on Wednesday, 16 November, at 17:00 SYNOPSIS Probabilistic programming is a popular tool for statistics and machine learning. The idea is to describe a statistical model as a program with random choices. The program might be a simulation of a system, such as a physics model, a model of viral spread, or a model of electoral behaviour. We can now carry out statistical inference over the system, for example, by running a Monte Carlo simulation – running the simulation 100,000’s of times. As I will discuss in this talk, the idea of treating statistical models as computer programs also has a foundational appeal. If we can understand statistical models as programs, then the foundations of probability and statistics can be discussed in terms of program semantics. There is a chance of new foundational perspectives on statistics, in terms of programming languages and their formal methods. As I will explain, this programming-based foundation for statistics is attractive because there are some intuitively simple scenarios, such as inference over function spaces, which have an easy programming implementation, but for which the traditional mathematical interpretation is complicated. SPEAKER BIOGRAPHY Sam Staton is a Professor of Computer Science and Royal Society University Research Fellow at the University of Oxford. There he currently runs an ERC grant "Better Languages for Statistics". Before arriving in Oxford in 2015, Sam spent time in Nijmegen, Paris, and Cambridge. His PhD was in Cambridge with Marcelo Fiore (2007). Sam's main research is in programming language theory, but he is also interested in logic and category theory. He has recent contributions in probabilistic programming languages, and quantum computing and programming languages. Webpage: https://www.cs.ox.ac.uk/people/samuel.staton/main.html [For admin and other information see: http://www.mta.ca/~cat-dist/ ] ^ permalink raw reply [flat|nested] 2+ messages in thread

end of thread, other threads:[~2022-11-14 12:49 UTC | newest]Thread overview:2+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2022-11-14 12:49 Sam Staton giving this year's LMS/BCS-FACS Evening Seminar (online) -- registration open until this Wednesday at 5PM, UTC Andrei Popescu -- strict thread matches above, loose matches on Subject: below -- 2022-11-14 12:49 Andrei Popescu

This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).