Continuous valued Q-learning method able to incrementally refine state space.

Masanori Takeda Takayuki Nakamura Tsukasa Ogasawara

Published in: IROS (2001)

Keyphrases

state space
dynamic programming
reinforcement learning
pairwise
detection algorithm
computational complexity
parameter estimation
non stationary
optimal policy
heuristic search
model free
wavelet analysis
discrete valued