Jakob Weissteiner

pic
Jakob Weissteiner
Ph.D. Candidate
Department of Informatics
University of Zurich
Binzmühlestrasse 14

CH-8050 Zürich

Website jakobweissteiner.com
Room

BIN 2.B.03

Tel +41 44 635 43 32
Email lastName[at]ifi[dot]uzh[dot]ch

 

Research Interests

Machine Learning, Deep Learning, Combinatorial Auctions, Market Design, Mechanism Design

Short Bio

Since February 2019, Jakob is a Ph.D. student advised by Prof. Sven Seuken in the Computation and Economics Research Group at the Department of Informatics of the University of Zurich, where he works on machine learning-based market design.

Jakob received a B.Sc. (2015) and a M.Sc. (2018) in Mathematics from the Technical University of Vienna (specialization: Financial and Actuarial Mathematics). Additionally he received a M.Sc. (2018) in Quantitative Finance from the Vienna University of Economics and Business.

Since September 2021 he is a ETH AI Center affiliated PhD student. He is currently as Workflow Chair part of the organizing committee of the twenty-third ACM Conference on Economics and Computation (EC'22).

Besides his studies, Jakob worked in the Advanced Analytics team of the Raiffeisen Bank International in Vienna. From September 2019 until March 2022 , he was a board member of the Club Alpbach Zürich.

Publications

  1. Deep Learning-powered Iterative Combinatorial Auctions. 
    Jakob Weissteiner and Sven Seuken.
    In Proceedings of the Thirty-fourth AAAI Conference on Artificial Intelligence (AAAI '20), New York, USA, February 2020.
    Working paper version from Oct 2020: [pdf] [code]
  2. Fourier Analysis-based Iterative Combinatorial Auctions.
    Jakob Weissteiner*, Chris Wendler*, Sven Seuken, Ben Lubin, and Markus Püschel.
    In Proceedings of the Thirty-first International joint Conference on Artificial Intelligence (IJCAI '22), Vienna, AUT, July 2022.
    Full paper version including appendix:  [pdf] [code]
  3. Monotone-Value Neural Networks: Exploiting Preference Monotonicity in Combinatorial Assignment
    Jakob Weissteiner*, Jakob Heiss*, Julien Siems* and Sven Seuken.
    In Proceedings of the Thirty-first International joint Conference on Artificial Intelligence (IJCAI '22), Vienna, AUT, July 2022.
    Full paper version including appendix: [pdf] [code]
  4. NOMU: Neural Optimization-based Model Uncertainty
    Jakob Weissteiner*, Hanna Wutte*, Jakob Heiss*, Sven Seuken, and Josef Teichmann.
    In Proceedings of the Thirty-ninth International Conference on Machine Learning (ICML '22), Washington, USA, July 2022.
    Full paper version including appendix:  [pdf] [code]

*These authors contributed equally

Working Papers

  1. Machine Learning-powered Course Allocation.
    Ermis Soumalias*, Behnoosh Zamanlooy*, Jakob Weissteiner and Sven Seuken.
    Preprint available upon request.

Master's Theses

  1. Variable importance measures in classification and regression methods. 
    Jakob Weissteiner. Master's Thesis. Vienna University of Economics and Business, Austria, Sep 2018. [pdf] (PDF, 1 MB)
  2. Über die Orderbuchmodellierung mit Markovschen Ketten in stetiger Zeit.
    Jakob Weissteiner. Master's Thesis. Technical University of Vienna, Austria, Jan 2018. [pdf] (PDF, 2 MB)

Curriculum Vitae

Teaching

2021: Head teaching assistant for lecture Market Design and Machine Learning 
          Head teaching assistant for lecture Seminar: Advanced Topics in Economics and Computation

2020: Head teaching assistant for lecture Seminar: Advanced Topics in Economics and Computation

2019: Teaching assistant for lecture Economics and Computation
          Head teaching assistant for Seminar: Advanced Topics in Economics and Computation.

2018: Teaching assistant for lecture Economics and Computation
          Teaching assistant for lecture Mathematics II, Vienna University of Economics and Business.

2017: Teaching assistant for lecture Probability Theory, Vienna University of Economics and Business.

2016: Teaching assistant for lecture Risk Management in Finance and Insurance, Technical University of Vienna.

Advised Theses

  • Bayesian Optimization with Neural Networks
    Master Thesis by Marius Högger, Fall 2020