Science ❯ Artificial Intelligence ❯ Machine Learning ❯ Reinforcement Learning
The open-source LATENT method builds a correctable action space from brief motion-capture fragments to transfer robust policies from simulation to the court.
