2917 search results - page 112 / 584 » Adaptation and the effort needed to adapt |
148
click to vote
JAT
2006
15 years 6 months ago
2006
Opposed to linear schemes, nonlinear function approximation allows to obtain a dimension independent rate of convergence. Unfortunately, in the presence of data noise typical algo...
152
click to vote
KI
2002
Springer
15 years 6 months ago
2002
Springer
In this paper, we present an approach for learning interest profiles implicitly from positive user observations only. This approach eliminates the need to prompt users for ratings...
176
click to vote
SASO
2010
IEEE
15 years 4 months ago
2010
IEEE
In an open distributed system, computational resources are peer-owned, and distributed over time and space. The fact that these resources can dynamically join or leave the system (...
184
click to vote
HCI
2009
15 years 4 months ago
2009
This paper proposes a support tool for designers who have realized the potential benefits of using a scenario-based approach, yet need a more concrete guidance for its implementati...
177
click to vote
CDC
2010
IEEE
15 years 1 months ago
2010
IEEE
Abstract-- We consider reinforcement learning, and in particular, the Q-learning algorithm in large state and action spaces. In order to cope with the size of the spaces, a functio...