Sciweavers

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

12

ICML
2005
IEEE

favoriteEmaildiscussreport

157views Machine Learning» more ICML 2005»

A causal approach to hierarchical decomposition of factored MDPs

14 years 5 months ago

A causal approach to hierarchical decomposition of factored MDPs

Download www-anw.cs.umass.edu

We present Variable Influence Structure Analysis, an algorithm that dynamically performs hierarchical decomposition of factored Markov decision processes. Our algorithm determines causal relationships between state variables and introduces temporallyextended actions that cause the values of state variables to change. Each temporallyextended action corresponds to a subtask that is significantly easier to solve than the overall task. Results from experiments show great promise in scaling to larger tasks.

Anders Jonsson, Andrew G. Barto

Real-time Traffic

Determines Causal Relationships | Factored Markov Decision | ICML 2005 | Machine Learning | State Variables |

claim paper

Related Content

» Causal Graph Based Decomposition of Factored MDPs

» TeXDYNA Hierarchical Reinforcement Learning in Factored MDPs

» Learning stateaction basis functions for hierarchical MDPs

» Generating Hierarchical Structure in Reinforcement Learning from State Variables

» Defining factors goals and criteria for reusable component evaluation

» Hierarchical multilinear models for multiway data

» Hierarchical Representations with Signatures for Large Expression Management

» The Fast Downward Planning System

» Semanticmeaningful contentbased image retrieval in wavelet domain

Post Info
More Details (n/a)

Added	17 Nov 2009
Updated	17 Nov 2009
Type	Conference
Year	2005
Where	ICML
Authors	Anders Jonsson, Andrew G. Barto

Comments (0)