Learning Distributed Equilibria in Linear-Quadratic Stochastic Differential Games: An alpha-Potential Approach Speaker 1: Philip Plank Abstract: This talk analyses independent policy-gradient learning in N-player linear-quadratic stochastic differential games. Each player employs a distributed policy that depends only on its own state and updates its policy independently by following the gradient of its own objective. We show that the game admits an alpha-potential structure, where alpha is determined by the degree of pairwise interaction asymmetry. Exploiting this structure, we establish global linear convergence of these methods to an approximate equilibrium, with suboptimality proportional to the degree of asymmetry and complexity scaling approximately linearly with the population size. We further construct affine distributed equilibria beyond the mean field setting for both symmetric and asymmetric interactions. Based on joint work with Yufei Zhang (Imperial College London). Stochastic search with dynamic redundancy and mortality Speaker 2: Dr Samantha Linn Abstract: Stochastic search is ubiquitous in physical and artificial realms. In these settings, the number of ‘searchers’ is rarely constant: new agents may be recruited while others can abandon the search. Despite the prevalence of these dynamics, their combined influence on search times remains largely unexplored. In this talk we will introduce frameworks for studying stochastic search in which agents progressively join and leave the process, a mechanism we term ‘dynamic redundancy and mortality’. Under minimal assumptions on the underlying search dynamics, we determine the exact distribution of the first-passage time to a target region and reveal surprising connections to stochastic search with stochastic resetting, wherein a single searcher is randomly ‘reset’ to its initial state. We will then treat the target region as a queue, which we show has interarrival times governed by a nonhomogeneous Poisson pr

Description supplied by the event organiser.