PeopleofSuperintelligence

Researcher

Robert Peharz

Robert Peharz is a machine learning researcher at Graz University of Technology whose work on sum-product networks and probabilistic circuits targets models that permit exact and efficient inference. After a master's degree and a PhD at TU Graz, finished in 2015, he worked as a postdoc at the Medical University of Graz and at Cambridge's Machine Learning Group, where he also held a Marie Curie Individual Fellowship. He joined Eindhoven University of Technology as an assistant professor in November 2019, introduced Einsum Networks at ICML in 2020, and returned to TU Graz in 2021. In July 2025, a paper with Johannes Exenberger and Sascha Ranftl, Deep Polynomial Chaos Expansion, won the Best Paper Award at the Workshop on Tractable Probabilistic Modeling at UAI. Promoted in 2026, he is an associate professor at TU Graz and Speaker of the Graz Center for Machine Learning.

Now at
Graz University of Technology, Institute of Machine Learning and Neural Computation, Graz Center for Machine Learning and European Laboratory for Learning and Intelligent Systems
Badges

Timeline

  1. 2010–2015

    Master's and PhD at TU Graz

    In 2010 he earns a master's degree in computer engineering from Graz University of Technology. His PhD there, finished in 2015, covers probabilistic graphical models and sum-product networks, with applications to signal processing.

  2. 2015–2017

    Postdoc at Medical University Graz

    From 2015 to 2017 he is a postdoc at the Medical University of Graz, working on early recognition of neural maldevelopment through behavioral neuroscience. In October 2015 he lectures at OFAI in Vienna on sum-product networks for probabilistic modeling. In 2016 he coauthors, with Robert Gens, Franz Pernkopf and Pedro Domingos, a paper on the latent variable interpretation in sum-product networks, accepted for TPAMI.

  3. 2017–2019

    Marie Curie fellow at Cambridge

    From 2017 to 2018 he is a postdoc in the Machine Learning Group at the University of Cambridge, and from 2018 to 2019 a Marie Curie Individual Fellow there. In 2018 he coauthors Probabilistic Deep Learning using Random Sum-Product Networks with Antonio Vergari, Zoubin Ghahramani and others; its RAT-SPNs predict about as well as deep neural networks, and the work appears at UAI in 2019.

  4. 2019–2021

    Assistant professor at Eindhoven

    Eindhoven University of Technology takes him on in November 2019 as an assistant professor in its Data and Artificial Intelligence cluster. He stays there until 2021.

  5. 2020

    The Einsum Networks paper

    In 2020 he is first author of Einsum Networks at ICML, an implementation design for probabilistic circuits that combines many arithmetic operations in one einsum operation, with speedups and memory savings of up to two orders of magnitude. At IJCAI, ECAI, ECML PKDD and AAAI in 2020 he also gives tutorials on probabilistic circuits, the models that permit exact and efficient inference.

  6. 2021–2023

    Back at TU Graz

    He re-joins TU Graz in 2021 as an assistant professor. In 2022 he serves as area chair at UAI and ECML/PKDD and gives a NeurIPS tutorial on probabilistic circuits.

  7. 2024–2025

    VENTUS and a best paper

    In September 2024 he becomes consortium coordinator of VENTUS, a project funded by the Austrian Research Promotion Agency that applies causal, probabilistic and physics-informed machine learning to diagnosis and predictive maintenance in wind turbines. In July 2025 he, Johannes Exenberger and Sascha Ranftl win the Best Paper Award at the Tractable Probabilistic Modeling workshop at UAI 2025 for Deep Polynomial Chaos Expansion.

  8. Since 2026

    Associate Professor at TU Graz

    TU Graz promotes him to Associate Professor in 2026. At the Institute of Machine Learning and Neural Computation he is principal investigator of the Probabilistic Machine Learning Group and Speaker of the Graz Center for Machine Learning, and ELLIS lists him as a Scholar in its Graz unit.

Career

  1. Since 2026Institute of Machine Learning and Neural ComputationPrincipal Investigator
  2. Since 2026Graz Center for Machine LearningSpeaker
  3. Since 2026European Laboratory for Learning and Intelligent SystemsScholar
  4. Since 2021Graz University of TechnologyAssistant Professor, Associate Professor
  5. 2019–2021Eindhoven University of TechnologyAssistant Professor
  6. 2017–2019University of CambridgePostdoc, Marie Curie Individual Fellow
  7. 2015–2017Medical University of GrazPostdoc

Education

  1. 2015Graz University of TechnologyPhD
  2. 2010Graz University of TechnologyMaster's degree

From the biography.

  1. Google has the highest combined h-index among AI organizationsIncludes European Laboratory for Learning and Intelligent Systems. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  2. 44% of AI career moves go from one university to anotherIncludes European Laboratory for Learning and Intelligent Systems. We estimate how 26,439 AI career moves link universities, companies, labs, nonprofits and government research centers, with decade counts, net flows and the top paths.
  3. 5 of the 25 people with the most-cited recent papers also lead on career citationsIncludes European Laboratory for Learning and Intelligent Systems. We compare citations to 2023–2025 papers with career citations and h-index, after correcting reviewed publication-date errors: who leads, when they started and where they work.
  4. LinkedIn (1,377) and X (1,193) lead social platforms across 1,935 people shaping AIIncludes European Laboratory for Learning and Intelligent Systems. We find social, code or video accounts for 1,935 of the 7,547 people shaping AI (26%), led by LinkedIn (1,377), X (1,193) and GitHub (557).

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Robert Peharz yet. Send us a free-licensed photo.

  1. Robert Peharz | UAI/e uai.win.tue.nl. Checked 22 October 2026.
  2. Dipl.-Ing. Dr. Robert Peharz, Medical University, Graz - OFAI ofai.at. Checked 22 October 2026.
  3. [1601.06180] On the Latent Variable Interpretation in Sum-Product Networks arxiv.org. Checked 22 October 2026.
  4. [1806.01910] Probabilistic Deep Learning using Random Sum-Product Networks arxiv.org. Checked 22 October 2026.
  5. Robert Peharz robert-peharz.github.io. Checked 22 October 2026.
  6. [2004.06231] Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic Circuits arxiv.org. Checked 22 October 2026.
  7. Probabilistic Circuits: Representations, Inference, Learning and Applications – Digital Ecai 2020 digital.ecai2020.eu. Checked 22 October 2026.
  8. Robert Peharz - Graz University of Technology tugraz.elsevierpure.com. Checked 22 October 2026.
  9. IML - Publications tugraz.at. Checked 22 October 2026.
  10. Robert Peharz | European Laboratory for Learning and Intelligent Systems ellis.eu. Checked 22 October 2026.