Spectral Approaches to Learning Predictive Representations

Spectral Approaches to Learning Predictive Representations

A central problem in artificial intelligence is to choose actions to maximize reward in a partially observable, uncertain environment. To do so, we must obtain an accurate environment model, and then plan to maximize reward. However, for complex domains, specifying a model by hand can be a time cons...

Full description

Bibliographic Details
Main Author: Boots, Byron
Format: Others
Published: Research Showcase @ CMU 2012
Subjects:
System Identification
Reinforcement Learning
Spectral Learning
Predictive State Representations
Kernel Methods
Artificial Intelligence and Robotics
Online Access:http://repository.cmu.edu/dissertations/131
http://repository.cmu.edu/cgi/viewcontent.cgi?article=1132&context=dissertations

Internet

http://repository.cmu.edu/dissertations/131
http://repository.cmu.edu/cgi/viewcontent.cgi?article=1132&context=dissertations

Similar Items