1 paper
Jeong Woon Lee, Kyoleen Kwak, Daeho Kim +1
Policies learned via continuous actor-critic methods often exhibit erratic, high-frequency oscillations, making them unsuitable for physical deployment. Current approaches attempt…