PHYSTAT-SBI 2024 - Simulation Based Inference in Fundamental Physics

Europe/Zurich
Max Planck Institute for Physics

Max Planck Institute for Physics

Boltzmannstr. 8 85748 Garching Germany
Lukas Alexander Heinrich (Technische Universitat Munchen (DE)), Louis Lyons (Imperial College (GB)), Olaf Behnke (Deutsches Elektronen-Synchrotron (DE)), Kyle Stuart Cranmer (University of Wisconsin Madison (US)), Michael Kagan (SLAC National Accelerator Laboratory (US)), Ann Lee (Carnegie Mellon University), François LANUSSE, Gilles Louppe, Mikael Kuusela (Carnegie Mellon University (US))
Description

Fueled by the recent advances of Machine Learning in the last decade, a new breed of techniques have been developed to tackle statistical inference problems for "likelihood-free" cases, where it is possible to sample from the data-generating process (i.e. via stochastic simulators) but a closed form evaluation of the density is intractable.

This group of methods is known as "simulation-based inference" (SBI) or "likelihood-free inference" (LFI) and will be the dedicated topic of this PHYSTAT Workshop taking place from May 15th - May 17th 2024 at the Max-Planck Institute for Physics (MPP) in Garching near Munich.

PHYSTAT https://phystat.github.io/Website/) is a long-running workshop series that brings together statisticians, machine learning researchers and physicists to discuss shared topics and foster collaboration among the research communities.

Confirmed Invited Speakers:

  • Kyle Cranmer (U Wisconsin-Madison)
  • Antoine Wehenkel (Apple)
  • Gilles Louppe (U Liège)
  • Laurence Perreault-Levasseur (U Montréal)
  • Ann Lee (Carnegie Mellon)
  • Julia Linhart (INRIA)
  • Noemi Montel (U of Amsterdam)
  • Jakub Tomczak (Eindhoven)
  • Christoph Weniger (GRAPPA)
  • Alexander Held (U Wisconsin-Madison)
  • Paul Bürkner (TU Dortmund)
  • Francesca Capel (MPP Munich)
  • Mikael Kuusela (Carnegie Mellon)
  • Aishik Ghosh (UC Irvine)

 

 

(Credit: Axel Griesch/MPP)

Acknowledgement

We gratefully acknowledge support from:

Participants
Zoom Meeting ID
61486822942
Host
Lukas Alexander Heinrich
Useful links
Join via phone
Zoom URL
    • 9:00 AM 9:30 AM
      Breakfast Coffee & Arrival 30m
    • 9:30 AM 10:00 AM
      Welcome & Intro 30m
      Speaker: Lukas Alexander Heinrich (Technische Universitat Munchen (DE))
    • 10:00 AM 11:00 AM
      The frontiers of simulation-based inference (Part 1) 1h
      Speaker: Gilles Louppe
    • 11:00 AM 12:00 PM
      The frontiers of simulation-based inference (Part 2) 1h
      Speaker: Kyle Stuart Cranmer (University of Wisconsin Madison (US))
    • 2:00 PM 3:00 PM
      A Statistical Perspective on SBI 1h
      Speaker: Paul Bürkner
    • 3:00 PM 3:30 PM
      The Causal Chambers: Real Physical Systems as a Testbed for AI Methodology 30m
      Speaker: Juan L. Gamella
    • 3:30 PM 4:00 PM
      Coffee Break 30m
    • 4:00 PM 5:00 PM
      Diagnostics in SBI 1h
      Speaker: Julia Linhart
    • 5:00 PM 6:00 PM
      Simulation-based Inference in Astrophysics 1h
      Speaker: Dr Francesca Capel (Max Planck Institute for Physics)
    • 7:30 PM 9:30 PM
      Beergarden 2h
    • 9:00 AM 10:00 AM
      SBI in HEP Meets Reality 1h
      Speaker: Alexander Held (University of Wisconsin Madison (US))
    • 10:00 AM 10:30 AM
      Practical SBI for Measurements at the LHC 30m
      Speaker: Jay Ajitbhai Sandesara (University of Massachusetts (US))
    • 10:30 AM 11:00 AM
      Coffee Break 30m
    • 11:00 AM 11:30 AM
      Precision systematics for unbinned hypothesis tests 30m
      Speaker: Robert Schoefbeck (Austrian Academy of Sciences (AT))
    • 11:30 AM 12:00 PM
      Dataset-wide inference in the presence of systematic uncertainties 30m
      Speaker: Philipp Windischhofer (University of Chicago (US))
    • 2:00 PM 3:00 PM
      SBI Cosmo 1h
      Speaker: Laurence Perreault Levasseur
    • 3:00 PM 3:30 PM
      Robust Bayesian Inference for Simulator-based Models 30m
      Speaker: Harita Dellaporta
    • 3:30 PM 4:00 PM
      Coffee Break 30m
    • 4:00 PM 5:00 PM
      Model Misspecification and SBI 1h
      Speaker: Antoine Wehenkel
    • 5:00 PM 5:30 PM
      Goodness-of-Fit in SBI 30m
      Speaker: Noemi Anau Montel
    • 5:30 PM 6:00 PM
      PolySwyft: a sequential simulation-based nested sampler 30m
      Speaker: Dr Will Handley
    • 6:00 PM 7:00 PM
      Poster Session 1h
    • 9:00 AM 10:00 AM
      Enabling ML Techniques for SBI I 1h
      Speaker: Jakub Tomczak
    • 10:00 AM 10:30 AM
      Sourcerer: Sample-based Maximum Entropy Source Distribution Estimation 30m
      Speaker: Guy Moss
    • 10:30 AM 11:00 AM
      Coffee 30m
    • 11:00 AM 12:00 PM
      SBI Tools 1h
      Speaker: Christoph Weniger
    • 2:00 PM 3:00 PM
      Frequentist Topics in SBI 1h
      Speaker: Ann Lee (Carnegie Mellon University)
    • 3:00 PM 3:30 PM
      Classification under Nuisance Parameters and Generalized Label Shift in Likelihood-Free Inference 30m
      Speaker: Luca Masserano
    • 3:30 PM 4:00 PM
      Neural Simulation-Based Inference of the Neutron Star Equation of State directly from Telescope Spectra 30m
      Speaker: Len Brandes (Technical University of Munich)
    • 4:00 PM 4:30 PM
      Coffee Break 30m
    • 4:30 PM 5:00 PM
      Physics Impressions 30m
      Speaker: Aishik Ghosh (University of California Irvine (US))
    • 5:00 PM 5:30 PM
      Stats/ML Impressions 30m
      Speaker: Mikael Kuusela (Carnegie Mellon University (US))
    • 5:30 PM 6:00 PM
      Closeout 30m
      Speaker: Lukas Alexander Heinrich (Technische Universitat Munchen (DE))