T. Gebrekidan

Diagnosing Faults in Reinforcement Learning Simulators and World Models with Canonical Polynomial Invariants

A large literature builds physical structure into learned dynamics on the premise that models respecting the underlying physics predict better. We test that premise using exact polynomial invariants recovered from trajectories and canonicalised as reduced Gr\"obner bases over $\mathbb{Q}$. On Acrobot, exactness provide...