Adaptive State Space Formation in Reinforcement Learning.

Kazuyuki Samejima Takashi Omori

Published in: ICONIP (1998)

Keyphrases

state space
reinforcement learning
reinforcement learning algorithms
markov decision processes
optimal policy
adaptive control
heuristic search
action space
continuous state spaces

dynamic programming
partially observable
state variables
markov decision process
learning capabilities
transfer learning
dynamical systems
control problems
actor critic

function approximation
model free
machine learning
state abstraction
planning problems
search space
multi agent
temporal difference

finite state
sufficient conditions
markov chain
learning agent
learning algorithm
neural network
policy search
state and action spaces