ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
Despite popular belief, boosting algorithms and related coordinate descent methods are prone to overfitting. We derive modifications to AdaBoost and related gradient-based coordin...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
This paper addresses exact learning of Bayesian network structure from data and expert's knowledge based on score functions that are decomposable. First, it describes useful ...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
We develop a cyclical blockwise coordinate descent algorithm for the multi-task Lasso that efficiently solves problems with thousands of features and tasks. The main result shows ...
ICML
2009
IEEE
15 years 28 days 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 ...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
Group-Lasso estimators, useful in many applications, suffer from lack of meaningful variance estimates for regression coefficients. To overcome such problems, we propose a full Ba...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
The promise of unsupervised learning methods lies in their potential to use vast amounts of unlabeled data to learn complex, highly nonlinear models with millions of free paramete...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
We show how variational Bayesian inference can be implemented for very large generalized linear models. Our relaxation is proven to be a convex problem for any log-concave model. ...
ICML
2009
IEEE
15 years 28 days ago
2009
IEEE
We consider the exploration/exploitation problem in reinforcement learning (RL). The Bayesian approach to model-based RL offers an elegant solution to this problem, by considering...