Reinforcement learning under space and time constraints

Authors	H.H. van Seijen
Supervisors	F.C.A. Groen
Cosupervisors	S. Whiteson L.J.H.M. Kester
Award date	08-12-2011
ISBN	9789059863941
Number of pages	170
Organisations	Faculty of Science (FNWI) - Informatics Institute (IVI)
Document type	PhD thesis
Note	Research conducted at: Universiteit van Amsterdam
Language	English
Downloads	Thesis Cover Title pages Contents 1: Introduction 2: Background 3: Reinforcement learning under space and time constraints 4: Trading space and time for performance 5: Reducing the problem size by representation selection 6: Reducing the problem size by policy space reduction 7: Conclusions and future work A: Relationship between best-match LVM and TD(λ) B: Off-policy Monte Carlo update C: Proofs Publications by the author Bibliography Summary Samenvatting Acknowledgements Stellingen
Permalink to this page

Back

UvA-DARE