Sciweavers

ICML
2008
IEEE
14 years 11 months ago
The projectron: a bounded kernel-based Perceptron
We present a discriminative online algorithm with a bounded memory growth, which is based on the kernel-based Perceptron. Generally, the required memory of the kernelbased Percept...
Francesco Orabona, Joseph Keshet, Barbara Caputo
ICML
2008
IEEE
14 years 11 months ago
Fast nearest neighbor retrieval for bregman divergences
We present a data structure enabling efficient nearest neighbor (NN) retrieval for bregman divergences. The family of bregman divergences includes many popular dissimilarity measu...
Lawrence Cayton
ICML
2008
IEEE
14 years 11 months ago
Estimating labels from label proportions
Consider the following problem: given sets of unlabeled observations, each set with known label proportions, predict the labels of another set of observations, also with known lab...
Alex J. Smola, Novi Quadrianto, Quoc V. Le, Tib&ea...
ICML
2008
IEEE
14 years 11 months ago
Adaptive p-posterior mixture-model kernels for multiple instance learning
In multiple instance learning (MIL), how the instances determine the bag-labels is an essential issue, both algorithmically and intrinsically. In this paper, we show that the mech...
Hua-Yan Wang, Qiang Yang, Hongbin Zha
ICML
2008
IEEE
14 years 11 months ago
Metric embedding for kernel classification rules
In this paper, we consider a smoothing kernelbased classification rule and propose an algorithm for optimizing the performance of the rule by learning the bandwidth of the smoothi...
Bharath K. Sriperumbudur, Omer A. Lang, Gert R. G....
ICML
2008
IEEE
14 years 11 months ago
Training structural SVMs when exact inference is intractable
While discriminative training (e.g., CRF, structural SVM) holds much promise for machine translation, image segmentation, and clustering, the complex inference these applications ...
Thomas Finley, Thorsten Joachims
ICML
2008
IEEE
14 years 11 months ago
Manifold alignment using Procrustes analysis
In this paper we introduce a novel approach to manifold alignment, based on Procrustes analysis. Our approach differs from "semisupervised alignment" in that it results ...
Chang Wang, Sridhar Mahadevan
ICML
2008
IEEE
14 years 11 months ago
Statistical models for partial membership
We present a principled Bayesian framework for modeling partial memberships of data points to clusters. Unlike a standard mixture model which assumes that each data point belongs ...
Katherine A. Heller, Sinead Williamson, Zoubin Gha...
ICML
2008
IEEE
14 years 11 months ago
Fully distributed EM for very large datasets
In EM and related algorithms, E-step computations distribute easily, because data items are independent given parameters. For very large data sets, however, even storing all of th...
Jason Wolfe, Aria Haghighi, Dan Klein