We show how nonlinear embedding algorithms popular for use with shallow semisupervised learning techniques such as kernel methods can be applied to deep multilayer architectures, ...
Recently, several learning algorithms relying on models with deep architectures have been proposed. Though they have demonstrated impressive performance, to date, they have only b...
Hugo Larochelle, Dumitru Erhan, Aaron C. Courville...
We show how to use unlabeled data and a deep belief net (DBN) to learn a good covariance kernel for a Gaussian process. We first learn a deep generative model of the unlabeled da...
We propose a privacy-preserving protocol for computing aggregation queries over the join of private tables. In this problem, several parties wish to share aggregated information ov...
—We demonstrate DPCube, a component in our Health Information DE-identification (HIDE) framework, for releasing differentially private data cubes (or multi-dimensional histogram...