Most models of utility elicitation in decision support and interactive optimization assume a predefined set of "catalog" features over which user preferences are express...
Training principles for unsupervised learning are often derived from motivations that appear to be independent of supervised learning. In this paper we present a simple unificatio...
We show a close relationship between the Expectation - Maximization (EM) algorithm and direct optimization algorithms such as gradientbased methods for parameter learning. We iden...
Ruslan Salakhutdinov, Sam T. Roweis, Zoubin Ghahra...
We present a statistical optimization framework for solving the end-to-end problem of progressive transmission of images over noisy channels. We consider the impacts of transmissi...
iSCSI has a problem of drastic performance deterioration in the case of longer-latency. Thus, we have optimized iSCSI remote storage access through multiple layers. As a result, w...