Skip to Main content Skip to Navigation
Journal articles

Riemannian game dynamics

Abstract : We study a class of evolutionary game dynamics defined by bal- ancing a gain determined by the game’s payoffs against a cost of motion that captures the difficulty with which the population moves between states. Costs of motion are represented by a Riemannian metric, i.e., a state-dependent inner product on the set of population states. The replicator dynamics and the (Euclidean) projection dynamics are the archetypal examples of the class we study. Like these representative dynamics, all Riemannian game dynamics satisfy certain basic desiderata, including positive correlation and global convergence in potential games. Moreover, when the underlying Riemannian metric satisfies a Hessian integrability condition, the resulting dynamics pre- serve many further properties of the replicator and projection dynamics. We examine the close connections between Hessian game dynamics and reinforce- ment learning in normal form games, extending and elucidating a well-known link between the replicator dynamics and exponential reinforcement learning.
Document type :
Journal articles
Complete list of metadatas

Cited literature [67 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01382281
Contributor : Panayotis Mertikopoulos <>
Submitted on : Tuesday, October 9, 2018 - 3:31:22 PM
Last modification on : Friday, July 3, 2020 - 4:52:51 PM

File

GeometricDynamics.pdf
Files produced by the author(s)

Identifiers

Citation

Panayotis Mertikopoulos, William H. Sandholm. Riemannian game dynamics. Journal of Economic Theory, Elsevier, 2018, 177, pp.315-364. ⟨10.1016/j.jet.2018.06.002⟩. ⟨hal-01382281⟩

Share

Metrics

Record views

273

Files downloads

220