207
click to vote
JMLR
2012
13 years 9 months ago
2012
We transform the outputs of each hidden neuron in a multi-layer perceptron network to have zero output and zero slope on average, and use separate shortcut connections to model th...
202
Voted
JMLR
2012
13 years 9 months ago
2012
We propose a novel mixtures of Gaussian processes model in which the gating function is interconnected with a probabilistic logical model, in our case Markov logic networks. In th...
247
click to vote
JMLR
2012
13 years 9 months ago
2012
We present a unifying framework for information theoretic feature selection, bringing almost two decades of research on heuristic filter criteria under a single theoretical inter...
218
click to vote
JMLR
2012
13 years 9 months ago
2012
We introduce the graphlet decomposition of a weighted network, which encodes a notion of social information based on social structure. We develop a scalable algorithm, which combi...
200
click to vote
JMLR
2012
13 years 9 months ago
2012
We propose a new method for comparing learning algorithms on multiple tasks which is based on a novel non-parametric test that we call the Poisson binomial test. The key aspect of...
193
click to vote
JMLR
2012
13 years 9 months ago
2012
We consider the multiple-response regression problem, where the response is subject to sparse gross errors, in the high-dimensional setup. We propose a tractable regularized M-est...
216
click to vote
JMLR
2012
13 years 9 months ago
2012
This paper proposes a new randomized strategy for adaptive MCMC using Bayesian optimization. This approach applies to nondifferentiable objective functions and trades off explor...
196
click to vote
JMLR
2012
13 years 9 months ago
2012
We describe a novel approach to quadrature for ratios of probabilistic integrals, such as are used to compute posterior probabilities. This approach offers performance superior t...
239
click to vote
JMLR
2012
13 years 9 months ago
2012
The expectation maximization (EM) algorithm is a popular algorithm for parameter estimation in models with hidden variables. However, the algorithm has several non-trivial limitat...