JMLR
2010
13 years 9 days ago
2010
Hartigan's method for k-means clustering is the following greedy heuristic: select a point, and optimally reassign it. This paper develops two other formulations of the heuri...
JMLR
2010
13 years 9 days ago
2010
JMLR
2010
13 years 9 days ago
2010
Deep belief networks are a powerful way to model complex probability distributions. However, it is difficult to learn the structure of a belief network, particularly one with hidd...
JMLR
2010
13 years 9 days ago
2010
JMLR
2010
13 years 9 days ago
2010
We propose a non-linear graphical model for structured prediction. It combines the power of deep neural networks to extract high level features with the graphical framework of Mar...
JMLR
2010
13 years 9 days ago
2010
A new approach to regression regularization called the Pairwise Elastic Net is proposed. Like the Elastic Net, it simultaneously performs automatic variable selection and continuo...
JMLR
2010
13 years 9 days ago
2010
In this paper, we develop a novel online algorithm based on the Sequential Monte Carlo (SMC) samplers framework for posterior inference in Dirichlet Process Mixtures (DPM) (DelMor...
JMLR
2010
13 years 9 days ago
2010
Deep belief nets have been successful in modeling handwritten characters, but it has proved more difficult to apply them to real images. The problem lies in the restricted Boltzma...
JMLR
2010
13 years 9 days ago
2010
We study the problem of image denoising where images are assumed to be samples from low dimensional (sub)manifolds. We propose the algorithm of locally linear denoising. The algor...
JMLR
2010
13 years 9 days ago
2010