Many real-world problems are inherently hierarchically structured. The use of this structure in an agent’s policy may well be the key to improved scalability and higher performa...
In many settings, data is collected as multiple time series, where each recorded time series is an observation of some underlying dynamical process of interest. These observations...
John Cunningham, Zoubin Ghahramani, Carl Edward Ra...
This paper proposes a novel Bayesian approximation inference method for mixture modeling. Our key idea is to factorize marginal log-likelihood using a variational distribution ove...
In this paper we address the problem of pool based active learning, and provide an algorithm, called UPAL, that works by minimizing the unbiased estimator of the risk of a hypothe...
—We demonstrate DPCube, a component in our Health Information DE-identification (HIDE) framework, for releasing differentially private data cubes (or multi-dimensional histogram...