1800 search results - page 92 / 360 » Learning Restart Strategies |
144
click to vote
EWRL
2008
15 years 6 months ago
2008
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...
124
click to vote
COLING
2008
15 years 6 months ago
2008
Recent years have witnessed a growing interest in analogical learning for NLP applications. If the principle of analogical learning is quite simple, it does involve complex steps ...
117
click to vote
ICML
2009
IEEE
16 years 5 months ago
2009
IEEE
Humans and animals learn much better when the examples are not randomly presented but organized in a meaningful order which illustrates gradually more concepts, and gradually more ...
149
click to vote
GECCO
2010
Springer
15 years 9 months ago
2010
Springer
Derandomization by means of mirrored samples has been recently introduced to enhance the performances of (1, λ)and (1 + 2)-Evolution-Strategies (ESs) with the aim of designing fa...
161
click to vote
GECCO
2010
Springer
15 years 9 months ago
2010
Springer
The well-known Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is a robust stochastic search algorithm for optimizing functions defined on a continuous search space RD ....