Sciweavers

Share
EWRL
2008
9 years 3 days ago
Efficient Reinforcement Learning in Parameterized Models: Discrete Parameter Case
We consider reinforcement learning in the parameterized setup, where the model is known to belong to a parameterized family of Markov Decision Processes (MDPs). We further impose ...
Kirill Dyagilev, Shie Mannor, Nahum Shimkin
EWRL
2008
9 years 3 days ago
Variable Metric Reinforcement Learning Methods Applied to the Noisy Mountain Car Problem
Two variable metric reinforcement learning methods, the natural actor-critic algorithm and the covariance matrix adaptation evolution strategy, are compared on a conceptual level a...
Verena Heidrich-Meisner, Christian Igel
EWRL
2008
9 years 3 days ago
New Error Bounds for Approximations from Projected Linear Equations
We consider linear fixed point equations and their approximations by projection on a low dimensional subspace. We derive new bounds on the approximation error of the solution, whi...
Huizhen Yu, Dimitri P. Bertsekas
EWRL
2008
9 years 3 days ago
Probabilistic Inference for Fast Learning in Control
Carl Edward Rasmussen, Marc Peter Deisenroth
EWRL
2008
9 years 3 days ago
Markov Decision Processes with Arbitrary Reward Processes
Abstract. We consider a control problem where the decision maker interacts with a standard Markov decision process with the exception that the reward functions vary arbitrarily ove...
Jia Yuan Yu, Shie Mannor, Nahum Shimkin
EWRL
2008
9 years 3 days ago
Optimistic Planning of Deterministic Systems
If one possesses a model of a controlled deterministic system, then from any state, one may consider the set of all possible reachable states starting from that state and using any...
Jean-François Hren, Rémi Munos
EWRL
2008
9 years 3 days ago
Regularized Fitted Q-Iteration: Application to Planning
We consider planning in a Markovian decision problem, i.e., the problem of finding a good policy given access to a generative model of the environment. We propose to use fitted Q-i...
Amir Massoud Farahmand, Mohammad Ghavamzadeh, Csab...
EWRL
2008
9 years 3 days ago
Policy Learning - A Unified Perspective with Applications in Robotics
Policy Learning approaches are among the best suited methods for high-dimensional, continuous control systems such as anthropomorphic robot arms and humanoid robots. In this paper,...
Jan Peters, Jens Kober, Duy Nguyen-Tuong
books