Research Focus

My current research focuses on approximations and application of game theoretic concepts such as the shapley value and shapley interactions to the current field of explainable artificial intelligence. I work on developing new approximation algorithms for shapley interactions, which are both scalable to large feature counts and efficient. Particularly consistent estimator are of main interest.
Further I am also interested in preference learning and its connections to game theory.

Selected Publications

  • Santo M. A. R. Thies, Juan C. Alfaro, Viktor Bengs MORE-PLR: multi-output regression employed for partial label ranking Machine Learning 115, 18 (2026).
  • Santo M. A. R. Thies, Hubert Baniecki, R. Teal Witter, Eyke Hüllermeier, Maximilian Muschalik, Fabian Fumagalli Proxy-Based Approximation of Shapley and Banzhaf Interactions arXiv:2605.22738
  • Santo M. A. R. Thies, Viktor Bengs, Timo Kaufmann, Sebastian J. Vollmer, Eyke Hüllermeier Calibrated Preference Learning: The Case of Label Ranking arXiv:2605.30447