Multi-Armed Bandit and Reinforcement Learning Problems Eyal Even-Dar...  Home/Search
Context
Related
| jmlr.org/papers/volume...evendar06a.pdf Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: mit.edu/papers/v7/ (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: We incorporate statistical confidence intervals in both the multi-armed bandit and the reinforcement learning problems. In the bandit problem we show that given n arms, it suffices to pull the arms a total of O ) log(1/d) times to find an e-optimal arm with probability of at least 1-d. This bound matches the lower bound of Mannor and Tsitsiklis (2004) up to constants. We also devise action elimination procedures in reinforcement learning algorithms. We describe a framework that is... (Update)
Active bibliography (related documents): More All
0.5: Local Bandit Approximation for Optimal Learning Problems - Duff, Barto (Correct)
0.3: Estimated of Parameter Distributions for . . . - Dimitrakakis (2004) (Correct)
0.3: Nearly Optimal Exploration-Exploitation Decision Thresholds - Dimitrakakis (2006) (Correct)
Similar documents based on text:
0.0: Unknown - (Correct)
BibTeX entry: (Update)
@misc{ and-journal,
author = "Multi-Armed Bandit And",
title = "Journal of Machine Learning Research 7 (2006) 1079--1105 Submitted 2/05;
Published 6/06 Action Elimination and Stopping Conditions for the",
url = "citeseer.ist.psu.edu/760277.html" }
Citations (may not include all citations):658 Learning from Delayed Rewards (context) - Watkins - 1989
413 Neuro-Dynamic Programming (context) - Bertsekas, Tsitsiklis - 1995
413 Neuro-Dynamic Programming (context) - Bertsekas, Tsitsiklis - 1996
246 Markov Decision Processes (context) - Puterman - 1994
189 Reinforcement Learning (context) - Sutton, Barto - 1998
59 Gambling in a rigged casino: The adversarial multi-armed ban.. - Auer, Cesa-Bianchi et al. - 1995
29 Near-optimal reinforcement learning in polynomial time - Kearns, Singh - 1998
20 Some aspects of sequential design of experiments (context) - Robbins - 1952
17 An upper bound on the loss from approximate optimal-value fu.. - Singh, Yee - 1994
15 Chapman and Hall (context) - Berry, Fristedt - 1985
10 method for convergence of stochastic approximation and reinf.. (context) - Borkar, Meyn - 2000
10 The non-stochastic multi-armed bandit problem - Auer, Cesa-Bianchi et al. - 2002
8 Competitive queue policies for differentiated services - Aiello, Mansour et al. - 1979
7 Buffer overflow management in QoS switches - Kesselman, Lotker et al. - 1985
6 Approximately optimal approximate reinforcement learning - Kakade, Langford - 2002
4 Dynamic programming and Markov decision processes (context) - Howard - 1960
3 The sample complexity of exploration in the multi-armed band.. - Mannor, Tsitsiklis - 2002
2 A modified dynamic programming method for Markov decision pr.. (context) - MacQueen - 1966
1 Finite time bounds for sampling based fitted value iteration (context) - Szepesvri, Munos - 2005
1 Probability inequalities for sums of bounded random variable.. (context) - Even-Dar, Mansour et al. - 1963
Documents on the same site (http://jmlr.csail.mit.edu/papers/v7/): More
Estimating the "Wrong" Graphical Model: Benefits in the.. - Wainwright (2006) (Correct)
A Hierarchy of Support Vector Machines for Pattern Detection - Sahbi, Geman (2006) (Correct)
Efficient Learning of Label Ranking by Soft Projections.. - Shalev-Shwartz, Singer (2006) (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC