[ML Seminar] IDA ML Seminar - 10/06 - Martin Trapp - Probabilistic Machine Learning Meets Large-scale Deep Learning Models

[Louis Ohl]

Dear colleagues,

The IDA ML seminar is today at 13.30 in Alan Turing.

Best regards,
--
Louis

________________________________
Från: Louis Ohl <louis.ohl at liu.se>
Skickat: den 3 juni 2026 09:50
Till: idaint at ida.liu.se <idaint at ida.liu.se>; ml-seminars at lists.liu.se <ml-seminars at lists.liu.se>; ai-academy at groups.liu.se <ai-academy at groups.liu.se>
Ämne: IDA ML Seminar - 10/06 - Martin Trapp - Probabilistic Machine Learning Meets Large-scale Deep Learning Models

Welcome to an IDA Machine Learning Seminar on Wednesday, June 10 at 13:30 In Alan Turing

Title (tentative): Probabilistic Machine Learning Meets Large-scale Deep Learning Models

Speaker: Martin Trapp<https://trappmartin.github.io/>, KTH, Computing and Learning Systems

Abstract (tentative): Modern deep learning models, such as LLMs and agentic systems, suffer from overconfidence and lack well-calibrated epistemic uncertainties. However, they are on such an enormous scale that classical Bayesian methods either do not apply directly or are challenging to scale. In this talk, I will discuss how probabilistic machine learning methods, particularly Bayesian methods, provide a valuable and urgently needed toolset for large-scale deep learning models, but require careful treatment to obtain effective and efficient approaches that deliver meaningful improvements. I will draw on recent work to discuss how stochastic tokenisation improves the robustness of LLMs, how Bayesian frameworks can be meaningfully integrated into agentic AI systems, and how links to tractable probabilistic models enable us to quantify quantisation errors in deep learning models in a principled way.

Location: Alan Turing, https://link.mazemap.com/RDyxFvXH

---------------
The list of future seminars in the series is available at:
https://stima.gitlab-pages.liu.se/ml-seminars/
You can subscribe to the seminar series' calendar using this ics link: https://outlook.office365.com/owa/calendar/4d811ae47ce446f58d11a7c2f50a7ed8@ad.liu.se/0f5253d7bc7841248c71eb4c28eb2d668927992292494627279/calendar.ics


--
Louis Ohl
Postdoc
louis.ohl at liu.se

Linköping University
IDA / STIMA
B-Huset, 2A:452
Webpage<https://oshillou.github.io/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.liu.se/pipermail/ml-seminars/attachments/20260610/e3dc8ae9/attachment.htm>