This website uses third party cookies to improve your experience. If you continue browsing or close this notice, you will accept their use.

Optimization, learning and games ``Replicator dynamics: old and new’’

26 October 2017 at 12:01 PM - 1:00 PM

Room 207, Campus on Viale Romania, 32

Speaker: Sylvain Sorin, Université Pierre et Marie Curie

We will describe the unilateral version associated to the replicator dynamics and its connection to  on-line learning procedures and classical gradient  algorithms in discrete and continuous time. 
We will survey recent results on extensions of this dynamics: regularization functions and  variable weights, time average  and link to  best reply dynamics in games,   equilibria and variational inequalities, potential and dissipative games,  Hessian Riemannian metrics.