Back to Results
First PageMeta Content
Dynamic programming / Stochastic control / Markov decision process / Reinforcement learning / Markov chain / Valuation / Secretary problem / Statistics / Markov processes / Markov models


Better Be Lucky Than Good: Exceeding Expectations in MDP Evaluation Thomas Keller and Florian Geißer University of Freiburg Freiburg, Germany {tkeller,geisserf}@informatik.uni-freiburg.de
Add to Reading List

Document Date: 2015-03-07 08:48:57

Open Document

File Size: 246,52 KB

Share Result on Facebook

City

Berlin / Cambridge / New York / /

Company

Princeton University Press / MIT Press / Microsoft / AAAI Press / /

Country

Germany / United States / /

/

Facility

Florian Geißer University of Freiburg Freiburg / University of Basel / /

IndustryTerm

online auctions / potential applications / sponsored search / approximate algorithms / Online Knapsack Problem / Online Reinforcement Learning / Online Probabilistic / considered algorithms / Web Conference / e - commerce / search tree / search space / /

Organization

University of Basel / American Statistical Association / MIT / Artificial Copyright Intelligence / MDP Evaluation Thomas Keller and Florian Geißer University of Freiburg Freiburg / Princeton University / Association for the Advancement / /

Person

Andrey Kolobov / Scott Sanner / /

Position

MAND planner / rt / related Secretary / planner / Secretary / /

ProgrammingLanguage

R / /

ProvinceOrState

Mississippi / /

PublishedMedium

Journal of Machine Learning Research / Journal of the American Statistical Association / /

Technology

MDP algorithms / four approximate algorithms / resulting algorithm / THTS algorithm / artificial intelligence / proposed algorithms / Machine Learning / MTE algorithm / simulation / depicted Algorithm / /

URL

www.aaai.org / /

SocialTag