Explore a groundbreaking approach to reinforcement learning in this 30-minute video that delves into concurrent control scenarios where agents must think and act simultaneously. Discover how researchers reformulate Q-learning in continuous time, introduce concurrency, and then revert to discrete time to address real-world challenges like robotic control. Learn about the novel continuous-time formulation of Bellman equations and their delay-aware discretization, leading to a new class of approximate dynamic programming methods. Examine the application of this framework to simulated benchmark tasks and a large-scale robotic grasping problem, demonstrating the practical implications of "thinking while moving" in reinforcement learning.