MultiSTOP: Solving Functional Equations with Reinforcement Learning