% 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”.

@ARTICLE{Bunino:1052123,
      author       = {Bunino, Matteo and Sæther, Jarl Sondre and Eickhoff, Linus
                      Maximilian and Lappe, Anna Elisa and Tsolaki, Kalliopi and
                      Verder, Killian and Mutegeki, Henry and Machacek, Roman and
                      Girone, Maria and Krochak, Oleksandr and Rüttgers, Mario
                      and Sarma, Rakesh and Lintermann, Andreas},
      title        = {itwinai: {A} {P}ython {T}oolkit for {S}calable {S}cientific
                      {M}achine {L}earning on {HPC} {S}ystems},
      journal      = {The journal of open source software},
      volume       = {11},
      number       = {117},
      issn         = {2475-9066},
      address      = {[Erscheinungsort nicht ermittelbar]},
      publisher    = {[Verlag nicht ermittelbar]},
      reportid     = {FZJ-2026-00771},
      pages        = {9409},
      year         = {2026},
      abstract     = {The integration of Artificial Intelligence (AI) into
                      scientific research has expanded significantlyover the past
                      decade, driven by the availability of large-scale datasets
                      and Graphics ProcessingUnits (GPUs), in particular at High
                      Performance Computing (HPC) sites. <br>However, many
                      researchers face significant barriers when deploying AI
                      workflows on HPCsystems, as their heterogeneous nature
                      forces scientists to focus on low-level
                      implementationdetails rather than on their core research. At
                      the same time, the researchers often lackspecialized HPC/AI
                      knowledge to implement their workflows efficiently. <br>To
                      address this, we present itwinai, a Python library that
                      simplifies scalable AI on HPC. Itsmodular architecture and
                      standard interface allow users to scale workloads
                      efficiently fromlaptops to supercomputers, reducing
                      implementation overhead and improving resource usage.},
      cin          = {JSC},
      ddc          = {004},
      cid          = {I:(DE-Juel1)JSC-20090406},
      pnm          = {5111 - Domain-Specific Simulation $\&$ Data Life Cycle Labs
                      (SDLs) and Research Groups (POF4-511) / interTwin - An
                      interdisciplinary Digital Twin Engine for science
                      (101058386) / SDLFSE - SDL Fluids $\&$ Solids Engineering
                      (SDLFSE)},
      pid          = {G:(DE-HGF)POF4-5111 / G:(EU-Grant)101058386 /
                      G:(DE-Juel-1)SDLFSE},
      typ          = {PUB:(DE-HGF)16},
      doi          = {10.21105/joss.09409},
      url          = {https://juser.fz-juelich.de/record/1052123},
}