6 search results - page 1 / 2 » Transfer of policies based on trajectory libraries |
IROS
2007
IEEE
13 years 11 months ago
2007
IEEE
— Libraries of trajectories are a promising way of creating policies for difficult problems. However, often it is not desirable or even possible to create a new library for ever...
ICRA
2006
IEEE
13 years 11 months ago
2006
IEEE
— We present a control approach that uses a library of trajectories to establish a global control law or policy. This is an alternative to methods for finding global policies ba...
AAAI
2010
13 years 6 months ago
2010
Bayesian inference is an appealing approach for leveraging prior knowledge in reinforcement learning (RL). In this paper we describe an algorithm for discovering different classes...
ICMLA
2003
13 years 6 months ago
2003
This work represents the first step towards a task library system in the reinforcement learning domain. Task libraries could be useful in speeding up the learning of new tasks th...
ICRA
2010
IEEE
13 years 3 months ago
2010
IEEE
Abstract— We consider the problem of apprenticeship learning when the expert’s demonstration covers only a small part of a large state space. Inverse Reinforcement Learning (IR...