Unai Fischer Abaigar

profile-photo.jpg

Hello! I’m a PhD student in Statistics at the University of Munich (LMU), where I study the foundations of prediction systems used to guide resource allocations in the public sector. A core theme of my research is thinking holistically about decision-making systems, examining not just the predictive algorithms themselves but how they are designed, deployed, and integrated into broader decision-making processes.

I joined the Social Data Science and AI Lab in October 2022, supported by the Konrad Zuse School for Excellence in Reliable AI and the Munich Center for Machine Learning. I’m advised by Christoph Kern and Frauke Kreuter.

In fall 2024, I was a visiting fellow at Harvard’s John A. Paulson School of Engineering and Applied Sciences, hosted by Cynthia Dwork and Juan Carlos Perdomo.

I have a background in Physics with a Bachelor’s and Master’s degree from Heidelberg University, where I specialized in dynamical systems, networks, and statistical machine learning for time series. Before my PhD, I worked as a research associate at the Hertie School Data Science Lab in Berlin, where I explored machine learning applications in public policy and governance.

Selected Research

The Value of Prediction in Identifying the Worst-Off
UFA, Christoph Kern and Juan Carlos Perdomo
International Conference on Machine Learning, 2025 (Outstanding Paper Award). Highlight Track @ FORC 2025.

Algorithms for Reliable Decision-Making Need Causal Reasoning
Christoph Kern, UFA, Jonas Schweisthal, Dennis Frauen, Rayid Ghani, Stefan Feuerriegel, Mihaela van der Schaar, and Frauke Kreuter
Nature Computational Science, 2025

Bridging the Gap: Towards an Expanded Toolkit for AI-Driven Decision-Making in the Public Sector
UFA, Christoph Kern, Noam Barda, and Frauke Kreuter
Government Information Quarterly, 2024

Full publication list

Awards, Grants and Fellowships

  • LMU–NYU Research Cooperation. Funding awarded for the project Foundations of Statistical Prediction in the Public Sphere, supporting a research stay at NYU (~€10,000)
  • ICML 2025 Outstanding Paper Award; received the award for The Value of Prediction in Identifying the Worst-Off; one of only six papers selected from the main track at ICML 2025
  • Selected for Konrad Zuse School for Excellence in Reliable AI, a competitive program offering funding and support for research in reliable and socially responsible AI
  • Awarded a scholarship from the German Academic Scholarship Foundation (Studienstiftung des deutschen Volkes), Germany’s most prestigious academic foundation (top 0.5% of students)

Talks

  • The Value of Prediction in Identifying the Worst-Off, ICML 2025 (Oral), Vancouver (Canada), July 2025.
  • Adjusting Survey Estimates Using Multiaccuracy Postprocessing, ITACOSM 2025, Bologna (Italy), July 2025.
  • The Value of Prediction in Identifying the Worst-Off, FORC 2025, Stanford University (US), June 2025.
  • The Value of Prediction in Identifying the Worst-Off, Social Foundations of Computation, Max Planck Institute for Intelligent Systems, Tübingen (Germany), May 2025.
  • Algorithmic Decision-Making in the Public Sector, Theory of Computation Graduate Student Seminar, Harvard University (US), October 2024.
  • Introduction to Automated Decision-Making, Coleridge Initiative, University of Maryland (Virtual), October 2024.
  • Machine Learning for Reliable Decision-Making, Zuse Industry Workshop on Algorithmic Decision-Making (Virtual), April 2024.
  • Challenges for ML-Supported Decision-Making, Department of Statistics Seminar, LMU Munich (Germany), January 2024.

Teaching

At the Unviersity of Munich (LMU)

  • Advanced Methods in Social Statistics and Social Data Science, Graduate Course (Summer 2024, 2025), Co-Instructor
  • Machine Learning Meets Causality, Graduate Seminar (Winter 2023), Co-Instructor
  • Computational Social Science, Graduate Course (Winter 2022, 2023, 2025), Teaching Assistant

At the Ruprecht Karl University of Heidelberg

  • Machine Learning for Real-World Challenges, Graduate Seminar (Summer 2022), Teaching Assistant
  • Dynamical Systems Theory in Machine Learning, Graduate Course (Winter 2021), Teaching Assistant

Outreach

  • Co-Organizer, Data Science for Social Good Munich (2022 – 2024). Helped coordinate a paid two-month fellowship where aspiring data scientists work on real-world machine learning projects in the public sector.
  • Co-Organizer, DataFest Germany (2023, 2025), a national data analysis hackathon aimed at students from across Germany.