We propose a generative model for text and other collections of discrete data that generalizes or improves on several previous models including naive Bayes/unigram, mixture of unigrams [6], and… Expand

A method for optimizing control policies, with guaranteed monotonic improvement, by making several approximations to the theoretically-justified scheme, called Trust Region Policy Optimization (TRPO).Expand

A simple spectral clustering algorithm that can be implemented using a few lines of Matlab is presented, and tools from matrix perturbation theory are used to analyze the algorithm, and give conditions under which it can be expected to do well.Expand

We consider problems involving groups of data where each observation within a group is a draw from a mixture model and where it is desirable to share mixture components between groups. We assume that… Expand

Statistical applications in fields such as bioinformatics, information retrieval, speech processing, image processing and communications often involve large-scale models in which thousands or… Expand

A new Deep Adaptation Network (DAN) architecture is proposed, which generalizes deep convolutional neural network to the domain adaptation scenario and can learn transferable features with statistical guarantees, and can scale linearly by unbiased estimate of kernel embedding.Expand

This work addresses the large number of samples typically required and the difficulty of obtaining stable and steady improvement despite the nonstationarity of the incoming data by using value functions to substantially reduce the variance of policy gradient estimates at the cost of some bias.Expand

This paper shows how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques and leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.Expand

This paper presents an algorithm that, given examples of similar (and, if desired, dissimilar) pairs of points in �”n, learns a distance metric over ℝn that respects these relationships.Expand

The prediction error for adversarial examples (robust error) is decompose as the sum of the natural (classification) error and boundary error, and a differentiable upper bound is provided using the theory of classification-calibrated loss, which is shown to be the tightest possible upper bound uniform over all probability distributions and measurable predictors.Expand