Discounted Sampling Policy Gradient for Robot Multi-objective Visual Control.

Meng Xu Qingfu Zhang Jianping Wang

Published in: EMO (2021)

Keyphrases

multi objective
policy gradient
evolutionary algorithm
mobile robot
optimization algorithm
control system
average reward

optimal control
genetic algorithm
reinforcement learning
control method
control problems
path planning

objective function
dynamic environments
optimal policy
sample size
monte carlo
control strategy

function approximation
model selection
humanoid robot
approximation methods
robot arm
cost function