% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@MISC{Bazarova:1052328,
      author       = {Bazarova, Alina and Robledo, Jose Ignacio and Kesselheim,
                      Stefan},
      title        = {{I}ntroduction to {S}imulation {B}ased {I}nference:
                      enhancing synthetic models with {A}rtificial {I}ntelligence},
      reportid     = {FZJ-2026-00936},
      year         = {2025},
      abstract     = {n the world of the research fields drifting further apart
                      from one another and Artificial Intelligence (AI) tools
                      gaining increasingly more attention, methods which can bring
                      a number of seemingly disjoint fields together is of the
                      utmost importance. The proposed tutorial is sought to
                      provide researchers with an instrument to unify Bayesian
                      modelling, large-scale simulations, and AI methods while
                      integrating them into the HPC environment.While Bayesian
                      inference is widely used in the research community, as it
                      provides distributional estimates of model parameters and
                      allows to update the model by incorporating new data into
                      it, it often suffers from computationally intensive
                      processes and limited parallelization capabilities.
                      Simulation Based Inference (SBI) is a tool to tackle this
                      issue.SBI employs AI-based approximate Bayesian computation
                      to dramatically reduce inference times and generate reliable
                      estimates, even when observed data are sparse. This approach
                      enables any representative simulation model to inform
                      parameter constraints, yielding approximate posterior
                      distributions. Additionally, SBI facilitates workload
                      distribution across high-performance computing clusters,
                      further reducing runtime.This tutorial discusses theoretical
                      foundations and provides practical training for constructing
                      SBI frameworks tailored to specific models. Through provided
                      examples, participants will gain insight into various levels
                      of model granularity, ranging from a simple black box
                      approach to a highly customizable design, and develop the
                      skills to effectively manage HPC devices within a given
                      set-up. By participating in this tutorial, attendees will
                      gain an understanding of the principles of Simulation Based
                      Inference, learn how to apply this methodology in the
                      context of HPC in a variety of case scenarios, to evaluate
                      its potential and utility, and be encouraged to consider its
                      applicability to their own research projects.},
      month         = {Jun},
      date          = {2025-06-02},
      organization  = {Karlsruhe (Germany), 2 Jun 2025 - 2
                       Jun 2025},
      subtyp        = {Other},
      cin          = {JSC},
      cid          = {I:(DE-Juel1)JSC-20090406},
      pnm          = {5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs)
                      and Research Groups (POF4-511) / Helmholtz AI Consultant
                      Team FB Information (E54.303.11)},
      pid          = {G:(DE-HGF)POF4-5112 / G:(DE-Juel-1)E54.303.11},
      typ          = {PUB:(DE-HGF)17},
      url          = {https://juser.fz-juelich.de/record/1052328},
}