A comparison of plan-based and abstract MDP reward shaping.

Kyriakos Efthymiadis Daniel Kudenko

Published in: Connect. Sci. (2014)

Keyphrases

reward shaping
reinforcement learning
markov decision processes
markov decision problems
decision theoretic
markov decision process
reward function
initial state
decision making
search algorithm
state space
linear programming
monte carlo
reinforcement learning algorithms