Sciweavers

JMLR
2008
117views more  JMLR 2008»
14 years 1 days ago
Active Learning by Spherical Subdivision
We introduce a computationally feasible, "constructive" active learning method for binary classification. The learning algorithm is initially formulated for separable cl...
Falk-Florian Henrich, Klaus Obermayer
JMLR
2008
141views more  JMLR 2008»
14 years 1 days ago
A New Algorithm for Estimating the Effective Dimension-Reduction Subspace
The statistical problem of estimating the effective dimension-reduction (EDR) subspace in the multi-index regression model with deterministic design and additive noise is consider...
Arnak S. Dalalyan, Anatoly Juditsky, Vladimir Spok...
JMLR
2008
150views more  JMLR 2008»
14 years 1 days ago
Discriminative Learning of Max-Sum Classifiers
The max-sum classifier predicts n-tuple of labels from n-tuple of observable variables by maximizing a sum of quality functions defined over neighbouring pairs of labels and obser...
Vojtech Franc, Bogdan Savchynskyy
JMLR
2008
69views more  JMLR 2008»
14 years 1 days ago
Classification with a Reject Option using a Hinge Loss
Peter L. Bartlett, Marten H. Wegkamp
JMLR
2008
139views more  JMLR 2008»
14 years 1 days ago
Regularization on Graphs with Function-adapted Diffusion Processes
Harmonic analysis and diffusion on discrete data has been shown to lead to state-of-theart algorithms for machine learning tasks, especially in the context of semi-supervised and ...
Arthur D. Szlam, Mauro Maggioni, Ronald R. Coifman
JMLR
2008
120views more  JMLR 2008»
14 years 1 days ago
Causal Reasoning with Ancestral Graphs
Causal reasoning is primarily concerned with what would happen to a system under external interventions. In particular, we are often interested in predicting the probability distr...
Jiji Zhang
JMLR
2008
79views more  JMLR 2008»
14 years 1 days ago
Manifold Learning: The Price of Normalization
We analyze the performance of a class of manifold-learning algorithms that find their output by minimizing a quadratic form under some normalization constraints. This class consis...
Yair Goldberg, Alon Zakai, Dan Kushnir, Yaacov Rit...
JMLR
2008
89views more  JMLR 2008»
14 years 1 days ago
An Error Bound Based on a Worst Likely Assignment
This paper introduces a new PAC transductive error bound for classification. The method uses information from the training examples and inputs of working examples to develop a set...
Eric Bax, Augusto Callejas
JMLR
2008
94views more  JMLR 2008»
14 years 1 days ago
Using Markov Blankets for Causal Structure Learning
We show how a generic feature selection algorithm returning strongly relevant variables can be turned into a causal structure learning algorithm. We prove this under the Faithfuln...
Jean-Philippe Pellet, André Elisseeff
JMLR
2008
83views more  JMLR 2008»
14 years 1 days ago
Shark
Christian Igel, Verena Heidrich-Meisner, Tobias Gl...