Learning predictive state representations in dynamical systems without reset.

Britton Wolfe Michael R. James Satinder P. Singh

Published in: ICML (2005)

Keyphrases

predictive state representations
dynamical systems
stochastic systems
temporal difference
differential equations
state space
partially observable
partially observable markov decision processes
learning algorithm
monte carlo
function approximation
stochastic models
nonlinear dynamical systems
past observations