Tereza Horáčková

Tereza Horáčková is Director of the Stochastic Games, Dynamics and Strategic Control Division at the Institute for Advanced Dynamic Uncertainty, and one of the leading mathematical authorities on mean field game theory and its applications to strategic policy. She holds a PhD in Mathematics from Univerzita Karlova — Charles University Prague (Matematicko-fyzikální fakulta), where her doctoral research established well-posedness and regularity results for a class of degenerate coupled HJB–Fokker–Planck systems arising in stochastic differential games with a continuum of players. Her thesis resolved several open questions concerning the existence of classical solutions to these systems under non-standard growth conditions on the Hamiltonian, and introduced a novel monotonicity framework that has since become a standard reference in the MFG literature.

Following her doctorate, Horáčková held a postdoctoral fellowship at CEREMADE — the Centre de Recherche en Mathématiques de la Décision at Université Paris-Dauphine — the institution where mean field game theory was originally developed by Pierre-Louis Lions and Jean-Michel Lasry. Working in the direct intellectual lineage of the founders of the field, she made independent contributions to the long-time behaviour of MFG systems, establishing exponential convergence rates for the equilibrium distribution to its ergodic limit and deriving sharp estimates on the speed of propagation of chaos in N-player approximations of the mean field limit.

Horáčková's research spans the full mathematical depth of the field: the viscosity solution theory of Hamilton-Jacobi equations on Wasserstein space, the analysis of master equations for MFG systems with common noise, the construction of weak solution frameworks for singular and degenerate mean field interactions, and the design of provably convergent numerical schemes for high-dimensional coupled PDE systems. What distinguishes her work is its insistence on mathematical completeness — she does not deploy a framework without first establishing the existence, uniqueness, and stability of the objects it claims to compute. Her results set the analytic standard against which numerical and applied work in the field is measured.

At IADU, Horáčková directs the scientific programme of the Stochastic Games, Dynamics and Strategic Control Division, overseeing a team of fellows and associates whose work spans stochastic optimisation, differential game theory, reinforcement learning, and convex analysis. Her own current research focuses on mean field game models for large-population policy problems: the design of optimal regulatory mechanisms for populations of strategically interacting agents, the analysis of MFG equilibria under model uncertainty and ambiguity, and the construction of efficient solution algorithms for the master equation in settings relevant to sovereign policy institutions.