Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

10

NIPS
2001

favoriteEmaildiscussreport

121views Information Technology» more NIPS 2001»

Rates of Convergence of Performance Gradient Estimates Using Function Approximation and Bias in Reinforcement Learning

13 years 5 months ago

Rates of Convergence of Performance Gradient Estimates Using Function Approximation and Bias in Reinforcement Learning

Download books.nips.cc

We address two open theoretical questions in Policy Gradient Reinforcement Learning. The first concerns the efficacy of using function approximation to represent the state action value function, . Theory is presented showing that linear function approximation representations of can degrade the rate of convergence of performance gradient estimates by a factor of

Gregory Z. Grudic, Lyle H. Ungar

Real-time Traffic

Action Value Function | Function Approximation | NIPS 2001 | NIPS 2007 | Policy Gradient Reinforcement |

claim paper

Related Content

» Estimation and Approximation Bounds for GradientBased Reinforcement Learning

» Exploiting Multiple Secondary Reinforcers in Policy Gradient Reinforcement Learning

» Residual Algorithms Reinforcement Learning with Function Approximation

» Reinforcement Learning in POMDPs via Direct Gradient Ascent

» SignaltoNoise Ratio Analysis of Policy Gradient Algorithms

» TD0 Converges Provably Faster than the Residual Gradient Algorithm

» Exploiting locality of interactions using a policygradient approach in multiagent learning

» Variance Reduction Techniques for Gradient Estimates in Reinforcement Learning

» Incremental Natural ActorCritic Algorithms

Post Info
More Details (n/a)

Added	31 Oct 2010
Updated	31 Oct 2010
Type	Conference
Year	2001
Where	NIPS
Authors	Gregory Z. Grudic, Lyle H. Ungar

Comments (0)