Year: 2026
Type: preprint
Abstract: We give an algorithm for computing an $\eps$-approximate fixed point of any $d$-dimensional nonexpansion with respect to the $\ell_\infty$ norm in time $\log(1/\eps)^{O(\log^2 d)}$. At its core, our method uses a boosting argument to geometrically shrink a search region. Among other consequences, our algorithm implies that simple stochastic games can be solved in quasipolynomial time.
Source: arXiv (Cornell University)
Authors: Soleymani, Ashkan, Farina, Gabriele
Institution: Massachusetts Institute of Technology
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: CAREER: Pushing the Boundaries of Learning Dynamics and Equilibrium Computation in Games: Control, Complexity, and Nonlinear Optimization, Collaborative Research: RI: Bridging the gap between deep reinforcement learning and imperfect-information games: sound policy gradient methods, search, and testbeds
OpenAlex ID: W7223230721