BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//euroscipy-2025//speaker//YZAGH8
BEGIN:VTIMEZONE
TZID:Europe/Warsaw
BEGIN:DAYLIGHT
DTSTART:20240819T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T030000
RDATE:20251026T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T030000
RDATE:20260329T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Beyond Likelihoods: Bayesian Parameter Inference for Black-Box Sim
 ulators with sbi - Jan Boelts (Teusen)\, Maternus Herold
DTSTART;TZID=Europe/Warsaw:20250819T153000
DTEND;TZID=Europe/Warsaw:20250819T170000
DTSTAMP:20260909T184340Z
UID:pretalx-euroscipy-2025-MU9HAJ@pretalx.com
DESCRIPTION:Do you spend time tuning parameters for complex scientific sim
 ulators? Perhaps you use grid search or optimization to match parameters t
 o data. These find a best-fit set\, but often don't reveal your confidence
  or if other parameters also fit. This uncertainty is crucial for reliable
  conclusions.\nThis tutorial introduces Simulation-Based Inference (SBI)\,
  a modern technique tackling this challenge. Unlike traditional Bayesian i
 nference methods (like MCMC) that require mathematical likelihood function
 s\, SBI works directly with your simulator's outputs. Using recent advance
 s in probabilistic ML\, it estimates the probability distribution of param
 eter values consistent with your observations\, even for complex "black-bo
 x" simulators. It provides not just a single best guess\, but full paramet
 er distributions representing parameter uncertainties and potential intera
 ctions.\nIn this hands-on tutorial using the `sbi` Python package\, you'll
  learn the practical steps: setting up the problem\, running SBI for param
 eter distributions\, and checking result reliability. We will cover differ
 ent SBI techniques and how to apply them.\nIf you are a scientist or engin
 eer using Python for simulations\, or just interested in probabilistic inf
 erence methods\, this session is for you. You will learn to obtain more re
 liable and interpretable results by quantifying uncertainty and understand
 ing how parameters interact within your model.\n[Link to material](https:/
 /github.com/janfb/euroscipy-2025-sbi-tutorial)
LOCATION:Room 1.38 (Ground Floor)
URL:https://pretalx.com/euroscipy-2025/talk/MU9HAJ/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Pyro Meets SBI: Unlocking Hierarchical Bayesian Inference for Comp
 lex Simulators - Jan Boelts (Teusen)
DTSTART;TZID=Europe/Warsaw:20250820T114000
DTEND;TZID=Europe/Warsaw:20250820T120000
DTSTAMP:20260909T184340Z
UID:pretalx-euroscipy-2025-KCYYTF@pretalx.com
DESCRIPTION:This talk introduces a novel approach that bridges Simulation-
 Based Inference (SBI) and probabilistic programming languages like Pyro to
  enable simulation-based hierarchical Bayesian inference. SBI is used to p
 erform parameter inference for intractable simulation models\, while Pyro 
 facilitates efficient Bayesian inference with complex hierarchical structu
 res. We demonstrate how to integrate SBI-learned likelihoods into Pyro mod
 els\, allowing for hierarchical Bayesian analysis of simulation-based mode
 ls. Using the drift-diffusion model from decision-making research as an ex
 ample\, we showcase the potential of this combined approach for tackling r
 eal-world problems with complex simulation models and hierarchical data. P
 resentation by Jan Teusen\, implementation in the `sbi` package by Seth Ax
 en. \n\nSlides and Code are available at [github.com/janfb/pyro-meets-sbi]
 (https://github.com/janfb/pyro-meets-sbi).
LOCATION:Room 1.38 (Ground Floor)
URL:https://pretalx.com/euroscipy-2025/talk/KCYYTF/
END:VEVENT
END:VCALENDAR