<ul><li>Researchers have addressed the problem of learning to control an unknown nonlinear dynamical system through sequential interactions.</li><li>They focus on achieving fast sequential learning in continuous control problems where the system dynamics depend smoothly on unknown parameters.</li><li>The study demonstrates that fast sequential learning is attainable if the optimal control policy is persistently exciting.</li><li>Additionally, they derive a regret bound which grows with the square root of the number of interactions for non-persistently exciting optimal policies.</li></ul>

Logarithmic Regret for Nonlinear Control

Discover more