Back to Results
First PageMeta Content
Maximum likelihood / Kullback–Leibler divergence / M-estimator / Likelihood function / Reinforcement learning / Divergence / Multi-armed bandit / Marginal likelihood / Extremum estimator / Statistics / Estimation theory / Expectation–maximization algorithm


Expectation Maximization for Weakly Labeled Data Yuri Ivanov MIT Media Laboratory, 20 Ames St., E15-390, Cambridge, MA 02139, USA
Add to Reading List

Document Date: 2003-06-30 17:21:48


Open Document

File Size: 346,84 KB

Share Result on Facebook

City

Cambridge / Sabes / San Diego / New York / /

Company

Weakly Labeled Data Yuri Ivanov MIT Media Laboratory / MIT Press / Bruce Blumberg MIT Media Laboratory / Ames / Alex Pentland MIT Media Laboratory / /

Country

Jordan / United States / /

/

Facility

Georgia Institute of Technology / /

IndustryTerm

on-line search / reinforcement pursuit algorithm / bandit search algorithms / analytical solution / on-line version / gradient-estimating algorithms / online reinforcement learning algorithms / neural networks / online clustering / on-line algorithm / subsequent search / low energy states / on-line fashion / search algorithm / learning algorithm / energy / /

Organization

MIT AI Lab / MIT / Georgia Institute of Technology / /

Person

Aaron Bobick / Leslie Pack Kaelbling / Lab / /

/

Position

bandit model / e.g / Walker / Forward / player / /

ProvinceOrState

P. N. / California / Massachusetts / /

PublishedMedium

Machine Learning / /

Technology

learning algorithm / EM-type algorithm / gradient-estimating algorithms / REM algorithms / REINFORCE algorithm / artificial intelligence / search algorithm / on-line algorithm / clustering algorithm / annealed REM algorithm / Expectation Maximization algorithm / machine learning / reinforcement pursuit algorithm / bandit search algorithms / EM algorithm / online reinforcement learning algorithms / /

SocialTag