ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
We present an improved bound on the difference between training and test errors for voting classifiers. This improved averaging bound provides a theoretical justification for popu...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
We present conditional random fields, a framework for building probabilistic models to segment and label sequence data. Conditional random fields offer several advantages over hid...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
In this paper, we show how we can learn to select good words for a document title. We view the problem of selecting good title words for a document as a variant of an Information ...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
This is a survey of some theoretical results on boosting obtained from an analogous treatment of some regression and classi cation boosting algorithms. Some related papers include...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
We call data weakly labeled if it has no exact label but rather a numerical indication of correctness of the label "guessed" by the learning algorithm - a situation comm...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
The Bayesian framework is ideally suited for induction problems. The probability of observing xt at
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
Hard disk drive failures are rare but are often costly. The ability to predict failures is important to consumers, drive manufacturers, and computer system manufacturers alike. In...
ICML
2001
IEEE
15 years 15 days ago
2001
IEEE
Hierarchical reinforcement learning (RL) is a general framework which studies how to exploit the structure of actions and tasks to accelerate policy learning in large domains. Pri...