3668 search results - page 216 / 734 » Margin Distribution and Learning |
223
click to vote
AIMSA
2008
Springer
15 years 9 months ago
2008
Springer
We present a hybrid approach to Distributed Constraint Satisfaction which combines incomplete, fast, penalty-based local search with complete, slower systematic search. Thus, we pr...
202
click to vote
CORR
2002
Springer
15 years 7 months ago
2002
Springer
Mutual information is widely used in artificial intelligence, in a descriptive way, to measure the stochastic dependence of discrete random variables. In order to address question...
201
click to vote
ECML
2007
Springer
16 years 1 months ago
2007
Springer
We present Policy Gradient Actor-Critic (PGAC), a new model-free Reinforcement Learning (RL) method for creating limited-memory stochastic policies for Partially Observable Markov ...
215
click to vote
NIPS
2008
15 years 8 months ago
2008
Policy gradient (PG) reinforcement learning algorithms have strong (local) convergence guarantees, but their learning performance is typically limited by a large variance in the e...
165
click to vote
LEGE
2004
15 years 8 months ago
2004
In this paper we discuss the TestTool system as an established testing system model, the one that is being used in real educational settings and supports self-assessment as well as...