Policy Return: A New Method for Reducing the Number of Experimental Trials in Deep Reinforcement Learning.

Feng Liu Shuling Dai Yongjia Zhao

Published in: IEEE Access (2020)

Keyphrases

cost function
dynamic programming
high accuracy
significant improvement
computational cost
reinforcement learning
detection method
computational complexity
real world
similarity measure
preprocessing
experimental evaluation
probabilistic model
supervised learning
support vector machine svm