Eric P. Xing

TL;DR: In this article, the authors present a parallel Markov chain Monte Carlo (MCMC) algorithm in which subsets of data are processed independently, with very little communication, and prove that their algorithm generates asymptotically exact samples and empirically demonstrate its ability to parallelize burn-in and sampling.

TL;DR: A tree-guided group lasso is proposed for estimating structured sparsity under multi-response regression by employing a novel penalty function constructed from the tree, and a systematic weighting scheme for the overlapping groups in the tree-penalty is described.

TL;DR: A completely general framework for learning sparse nonparametric directed acyclic graphs (DAGs) from data is developed that can be applied to general nonlinear models, general differentiable loss functions, and generic black-box optimization routines.

TL;DR: FlexiFaCT provides a distributed, scalable method for decomposing matrices, tensors, and coupled data sets through stochastic gradient descent on a variety of objective functions.

TL;DR: In this article, a tree-guided group lasso is proposed for estimating structured sparsity under multi-response regression by employing a novel penalty function constructed from the tree, and a systematic weighting scheme for the overlapping groups in the tree-penalty such that each regression coefficient is penalized in a balanced manner despite the inhomogeneous multiplicity of group memberships of the regression coefficients due to overlap among groups.