Starting Date: Summer 2026
Prerequisites: Algorithms, Linear algebra and probabilities
Will results be assigned to University: No
Nash equilibrium is one of the most fundamental solution concepts in game theory, where we know its existence in any finite game. In this project, we are interested in the computation of a Nash equilibrium in the specific class of bimatrix zero-sum games.
We will study learning algorithms for computing (approximate) Nash equilibria in zero-sum games and we will make a comparison with the current state of art bibliography.
Depending on the outcome of the project, we can potentially submit the results for a publication to a conference in Theoretical Computer Science, Artificial Intelligence or Machine learning.