113 search results - page 16 / 23 » Learning Representation and Control in Continuous Markov Dec... |
ILP
2007
Springer
14 years 3 months ago
2007
Springer
Abstract. Many reinforcement learning domains are highly relational. While traditional temporal-difference methods can be applied to these domains, they are limited in their capaci...
IJCAI
2007
13 years 10 months ago
2007
A long-lived agent continually faces new tasks in its environment. Such an agent may be able to use knowledge learned in solving earlier tasks to produce candidate policies for it...
AAAI
2006
13 years 10 months ago
2006
Markov decision processes (MDPs) with discrete and continuous state and action components can be solved efficiently by hybrid approximate linear programming (HALP). The main idea ...
AIPS
1998
13 years 10 months ago
1998
Planning methods for deterministic planning problems traditionally exploit factored representations to encode the dynamics of problems in terms of a set of parameters, e.g., the l...
ICRA
2007
IEEE
14 years 3 months ago
2007
IEEE
— The least squares approach works efficiently in value function approximation, given appropriate basis functions. Because of its smoothness, the Gaussian kernel is a popular an...