Adapting to Conflict: Equilibrium structure and adaptive learning in harmonic games
Event details
| Date | 09.04.2026 |
| Hour | 11:15 › 12:00 |
| Speaker | Davide Legacci, PhD candidate in Computer Science at Université Grenoble Alpes, CNRS and Inria, |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract
We propose an adaptive, “parameter-agnostic” algorithm for learning in harmonic games, a class of finite games characterized by strictly opposed, conflicting interests. In contrast to zero-sum games—which do not always capture conflict, even between two players—harmonic games may exhibit highly non-convex equilibrium configurations. In particular, if the game’s ambient dimension is odd, we show that the set of totally mixed Nash equilibria is a (typically non-convex) real algebraic variety of dimension at least one, which extends all the way to the boundary of the game’s strategy space. Despite the difficulties incurred by this complicated equilibrium structure, we provide an extrapolation-based variant of “follow-the-regularized-leader” (FTRL) which converges to Nash equilibrium with order-optimal regret guarantees: O(1) in self-play, and O( sqrt(T) ) against arbitrary play.
Biography
Davide Legacci is a PhD candidate in Computer Science at Université Grenoble Alpes, CNRS and Inria, within the GHOST team, supervised by Panayotis Mertikopoulos and Bary S. R. Pradelski. His research lies at the intersection of game theory and online learning, with a focus on the geometric structure of competitive interactions and the design of no-regret learning dynamics for adversarial environments. His work has been presented at leading machine learning venues including ICML and NeurIPS. He is expected to defend his PhD in September 2026.
We propose an adaptive, “parameter-agnostic” algorithm for learning in harmonic games, a class of finite games characterized by strictly opposed, conflicting interests. In contrast to zero-sum games—which do not always capture conflict, even between two players—harmonic games may exhibit highly non-convex equilibrium configurations. In particular, if the game’s ambient dimension is odd, we show that the set of totally mixed Nash equilibria is a (typically non-convex) real algebraic variety of dimension at least one, which extends all the way to the boundary of the game’s strategy space. Despite the difficulties incurred by this complicated equilibrium structure, we provide an extrapolation-based variant of “follow-the-regularized-leader” (FTRL) which converges to Nash equilibrium with order-optimal regret guarantees: O(1) in self-play, and O( sqrt(T) ) against arbitrary play.
Biography
Davide Legacci is a PhD candidate in Computer Science at Université Grenoble Alpes, CNRS and Inria, within the GHOST team, supervised by Panayotis Mertikopoulos and Bary S. R. Pradelski. His research lies at the intersection of game theory and online learning, with a focus on the geometric structure of competitive interactions and the design of no-regret learning dynamics for adversarial environments. His work has been presented at leading machine learning venues including ICML and NeurIPS. He is expected to defend his PhD in September 2026.
Practical information
- General public
- Free
Organizer
- Prof. Maryam Kamgarpour