@Article{Supelec875,
author = {Matthieu Geist and Bruno Scherrer},
title = {Off-policy Learning with Eligibility Traces: A Survey},
journal = {Journal of Machine Learning Research (JMLR)},
year = {2014},
volume = {15},
pages = {289-333},
url = {http://jmlr.org/papers/v15/geist14a.html},
abstract = {In the framework of Markov Decision Processes, we consider linear off-policy learning, that is the problem of learning a linear approximation of the value function of some fixed policy from one trajectory possibly generated by some other policy. We briefly review on-policy learning algorithms of the literature (gradient-based and least-squares-based), adopting a unified algorithmic view. Then, we highlight a systematic approach for adapting them to off-policy learning with eligibility traces. This leads to some known algorithms – off-policy LSTD($\lambda$), LSPE($\lambda$), TD($\lambda$), TDC/GQ($\lambda$) – and suggests new extensions – off-policy FPKF($\lambda$), BRM($\lambda$), gBRM($\lambda$), GTD2($\lambda$). We describe a comprehensive algorithmic derivation of all algorithms in a recursive and memory-efficent form, discuss their known convergence properties and illustrate their relative empirical behavior on Garnet problems. Our experiments suggest that the most standard algorithms on and off-policy LSTD($\lambda$)/LSPE($\lambda$) – and TD($\lambda$) if the feature space dimension is too large for a leastsquares approach – perform the best.}
}