Reinforcement Learning with Deep Energy-Based Policies.

Tuomas Haarnoja Haoran Tang Pieter Abbeel Sergey Levine

Published in: CoRR (2017)

Keyphrases

reinforcement learning
optimal policy
policy search
control policies
markov decision process
markov decision processes
partially observable markov decision processes
reinforcement learning agents
reward function
cooperative multi agent systems

reinforcement learning algorithms
total reward
control policy
function approximation
fitted q iteration
temporal difference
markov decision problems
hierarchical reinforcement learning
dynamic programming
model free

policy gradient methods
state space
long run
approximate policy iteration
decision problems
state and action spaces
machine learning
infinite horizon
management policies
deep learning

state abstraction
search space
multi agent reinforcement learning
decision processes
action space
macro actions
sufficient conditions
average cost
partially observable
data sets