Combining Soft-Actor Critic with Cross-Entropy Method for Policy Search in Continuous Control.

Hieu Trung Nguyen Khang Tran Ngoc Hoang Luong

Published in: CEC (2022)

Keyphrases

cross entropy
policy search
dynamic programming
objective function
linear programming
maximum likelihood
optimization method
optimal control
reinforcement learning
control system
probabilistic model
support vector machine
basis functions
decision problems
evaluation function
step size