2011 search results - page 187 / 403 » Universal Reinforcement Learning |
129
Voted
AIIDE
2008
15 years 7 months ago
2008
We introduce the ALeRT (Action-dependent Learning Rates with Trends) algorithm that makes two modifications to the learning rate and one change to the exploration rate of traditio...
133
click to vote
ALT
2005
Springer
16 years 1 months ago
2005
Springer
We bound the future loss when predicting any (computably) stochastic sequence online. Solomonoff finitely bounded the total deviation of his universal predictor M from the true ...
111
click to vote
BMCBI
2007
15 years 5 months ago
2007
abstract Niels Landwehr1 , Taneli Mielik¨ainen2 , Lauri Eronen2 , Hannu Toivonen1,2 , and Heikki Mannila2 1 Machine Learning Lab, Dept. of Comp. Science, University of Freiburg, G...
169
click to vote
ICML
2000
IEEE
16 years 5 months ago
2000
IEEE
Eligibility traces have been shown to speed reinforcement learning, to make it more robust to hidden states, and to provide a link between Monte Carlo and temporal-difference meth...
148
Voted
ECML
2004
Springer
15 years 10 months ago
2004
Springer
Abstract. We present first experiments using Support Vector Regression as function approximator for an on-line, sarsa-like reinforcement learner. To overcome the batch nature of S...