Navigation auf


Department of Informatics Computation and Economics Research Group

ERC Starting Grant: Machine Learning-based Market Design

ERC Logo


Principal Investigator: Prof. Dr. Sven Seuken

Grant Amount: 1.375 Million Euro

Project Time Period: Dec 2018 - Dec 2023

Keywords: Market Design, Machine Learning, Artifical Intelligence, Algorithmic Game Theory, Auctions, Matching

Project Summary

Market designers study how to set the "rules of a marketplace" such that the market works well. However, markets are getting increasingly complex such that designing good market mechanisms "by hand" is often infeasible, in particular when certain design desiderata (such as efficiency, strategyproofness, or fairness) are in conflict with each other. Moreover, human agents are boundedly-rational: already in small domains, they are best modeled as having incomplete preferences, because they may only know a ranking or the values of their top choices. In combinatorial domains, the number of choices grows exponentially, such that it quickly becomes impossible for an agent to report its full valuation, even if it had complete preferences. In this ERC project, we aim to combine techniques from "machine learning" with "market design" to address these challenges.

First, we will develop a new, automated approach to design mechanisms with the help of machine learning (ML). In contrast to prior ML-based automated mechanism design work, we will work towards the design of ML algorithms that exploit regularities in the mechanism design space. Second, we will study the "design of machine learning-based mechanisms." These are mechanisms that use machine learning internally to achieve good efficiency, revenue, fairness, or incentives even when agents have incomplete knowledge about their own preferences.
ERC Logo

In addition to pushing the scientific boundaries of market design research, this ERC project will also have an immediate impact on practical market design. We will apply our techniques in two different settings: (1) for the design of combinatorial spectrum auctions, a multi-billion dollar domain; and (2) for the design of matching markets (e.g., school choice, refugee matching, adoption matching).

  1. Uncertainty Bounds for Neural Networks - An algorithmic Approach.
    Jakob Weissteiner, Hanna Wutte, Jakob Heiss, Sven Seuken, and Josef Teichmann. Working Paper. February 2021.
  2. Search and Matching for Adoption from Foster Care.
    Nils Olberg, Ludwig Dierks, Sven Seuken, Vincent Slaugh, and M. Utku Ünver. Working Paper. February 2021.
  3. Machine Learning-powered Iterative Combinatorial Auctions with Interval Bidding. [pdf]
    Manuel Beyeler, Gianluca Brero, Benjamin Lubin, and Sven Seuken. Working Paper. September 2020.
  4. Fourier Analysis-based Iterative Combinatorial Auctions. [pdf]
    Jakob Weissteiner, Chris Wendler, Sven Seuken, Ben Lubin, and Markus Püschel. Working Paper. September 2020.
  5. Deep Learning-powered Iterative Combinatorial Auctions. [pdf]
    Jakob Weissteiner and Sven Seuken. In Proceedings of the Thirty-fourth AAAI Conference on Artificial Intelligence (AAAI-20), New York, NY, February 2020.
  6. Machine Learning-powered Iterative Combinatorial Auctions. [pdf]
    Gianluca Brero, Benjamin Lubin, and Sven Seuken. Working Paper. November 2019.
  7. Enabling Trade-offs in Machine Learning-based Matching for Refugee Resettlement. [pdf]
    Nils Olberg and Sven Seuken. Working Paper. June 2019.
  8. Fast Iterative Combinatorial Auctions via Bayesian Learning. [pdf]
    Gianluca Brero, Sébastien Lahaie, and Sven Seuken. In Proceedings of the Thirty-third AAAI Conference of Artificial Intelligence (AAAI-19), Honolulu, USA, January 2019.
  9. Combinatorial Auctions via Machine Learning-based Preference Elicitation. [pdf]
    Gianluca Brero, Benjamin Lubin, and Sven Seuken. In Proceedings of the Twenty-seventh International Joint Conference on Artificial Intelligence and the Twenty-third European Conference on Artificial Intelligence (IJCAI-ECAI-18), Stockholm, Sweden, July 2018.
  10. Probably Approximately Efficient Combinatorial Auctions via Machine Learning. [pdf]
    Gianluca Brero, Benjamin Lubin, and Sven Seuken. In Proceedings of the Thirty-first AAAI Conference of Artificial Intelligence (AAAI-17), San Francisco, USA, February 2017.

In the News

  1. Interview with Sven Seuken in the UZH Magazin
    UZH Department of Informatics News Entry, March 24, 2020

  2. ERC Starting Grant 2018 for Sven Seuken
    UZH Department of Informatics News Entry, August 9, 2018

  3. Four Million Euros Awarded to Three UZH Researchers
    University of Zurich Press Release, August 9, 2018

  4. Forschungsprojekt zu maschinellem Lernen erhält EU-Fördergeld
    Horizont, August 9, 2018

  5. Vier Schweizer ICT-Projekte erhalten Millionenförderung der EU
    Netzwoche, August 14, 2018