DROPS

Document

Automated Machine Learning For Computational Mechanics (Dagstuhl Seminar 24282)

Authors: Elena Raponi, Lars Kotthoff, Hyunsun Alicia Kim, and Marius Lindauer

Published in: Dagstuhl Reports, Volume 14, Issue 7 (2025)

Abstract

Machine learning (ML) has achieved undeniable success in computational mechanics, an ever-growing discipline that impacts all areas of engineering, from structural and fluid dynamics to solid mechanics and vehicle simulation. Computational mechanics uses numerical models and time- and resource-consuming simulations to reproduce physical phenomena, usually with the goal of optimizing the parameter configuration of the model with respect to the desired properties of the system. ML algorithms enable the construction of surrogate models that approximate the outcome of the simulations, allowing faster identification of well-performing configurations. However, determining the best ML approach for a given task is not straightforward and depends on human experts. Automated machine learning (AutoML) aims to reduce the need for experts to obtain effective ML pipelines. It provides off-the-shelf solutions that can be used without prior knowledge of ML, allowing engineers to spend more time on domain-specific tasks. AutoML is underutilized in computational mechanics; there is almost no communication between the two communities, and engineers spend unnecessary effort selecting and configuring ML algorithms. Our Dagstuhl Seminar aimed to (i) raise awareness of AutoML in the computational mechanics community, (ii) discover strengths and challenges for applying AutoML in practice, and (iii) create a bilateral exchange so that researchers can mutually benefit from their complementary goals and needs.

Cite as

Elena Raponi, Lars Kotthoff, Hyunsun Alicia Kim, and Marius Lindauer. Automated Machine Learning For Computational Mechanics (Dagstuhl Seminar 24282). In Dagstuhl Reports, Volume 14, Issue 7, pp. 17-34, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{raponi_et_al:DagRep.14.7.17,
  author =	{Raponi, Elena and Kotthoff, Lars and Kim, Hyunsun Alicia and Lindauer, Marius},
  title =	{{Automated Machine Learning For Computational Mechanics (Dagstuhl Seminar 24282)}},
  pages =	{17--34},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2025},
  volume =	{14},
  number =	{7},
  editor =	{Raponi, Elena and Kotthoff, Lars and Kim, Hyunsun Alicia and Lindauer, Marius},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.14.7.17},
  URN =		{urn:nbn:de:0030-drops-229331},
  doi =		{10.4230/DagRep.14.7.17},
  annote =	{Keywords: automated algorithm design; computational mechanics; engineering applications of AI; black-box optimization; physics-informed machine learning}
}

Document

DOI: 10.4230/DagRep.13.9.116

Human-Centered Approaches for Provenance in Automated Data Science (Dagstuhl Seminar 23372)

Authors: Anamaria Crisan, Lars Kotthoff, Marc Streit, and Kai Xu

Published in: Dagstuhl Reports, Volume 13, Issue 9 (2024)

Abstract

The scope of automated machine learning (AutoML) technology has extended beyond its initial boundaries of model selection and hyperparameter tuning and towards end-to-end development and refinement of data science pipelines. These advances, both theoretical and realized, make the tools of data science more readily available to domain experts that rely on low- or no-code tooling options to analyze and make sense of their data. To ensure that automated data science technologies are applied both effectively and responsibly, it becomes increasingly urgent to carefully audit the decisions made both automatically and with guidance from humans. This Dagstuhl Seminar examines human-centered approaches for provenance in automated data science. While prior research concerning provenance and machine learning exists, it does not address the expanded scope of automated approaches and the consequences of applying such techniques at scale to the population of domain experts. In addition, most of the previous works focus on the automated part of this process, leaving a gap on the support for the sensemaking tasks users need to perform, such as selecting the datasets and candidate models and identifying potential causes for poor performance. The seminar brought together experts from across provenance, information visualization, visual analytics, machine learning, and human-computer interaction to articulate the user challenges posed by AutoML and automated data science, discuss the current state of the art, and propose directions for new research. More specifically, this seminar: - articulates the state of the art in AutoML and automated data science for supporting the provenance of decision making, - describes the challenges that data scientists and domain experts face when interfacing with automated approaches to make sense of an automated decision, - examines the interface between data-centric, model-centric, and user-centric models of provenance and how they interact with automated techniques, and - encourages exploration of human-centered approaches; for example leveraging visualization.

Cite as

Anamaria Crisan, Lars Kotthoff, Marc Streit, and Kai Xu. Human-Centered Approaches for Provenance in Automated Data Science (Dagstuhl Seminar 23372). In Dagstuhl Reports, Volume 13, Issue 9, pp. 116-136, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2024)

Copy BibTex To Clipboard

@Article{crisan_et_al:DagRep.13.9.116,
  author =	{Crisan, Anamaria and Kotthoff, Lars and Streit, Marc and Xu, Kai},
  title =	{{Human-Centered Approaches for Provenance in Automated Data Science (Dagstuhl Seminar 23372)}},
  pages =	{116--136},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2024},
  volume =	{13},
  number =	{9},
  editor =	{Crisan, Anamaria and Kotthoff, Lars and Streit, Marc and Xu, Kai},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.13.9.116},
  URN =		{urn:nbn:de:0030-drops-198236},
  doi =		{10.4230/DagRep.13.9.116},
  annote =	{Keywords: Dagstuhl Seminar, Provenance, AutoML, Data Science, Information Visualisation, Visual Analytics, Machine Learning, Human-Computer Interaction}
}

Search Results

Documents authored by Kotthoff, Lars

Automated Machine Learning For Computational Mechanics (Dagstuhl Seminar 24282)

Abstract

Cite as

Human-Centered Approaches for Provenance in Automated Data Science (Dagstuhl Seminar 23372)

Abstract

Cite as

Thanks for your feedback!

Could not send message