Multiple experts, each predicting a potential file successor, improve accuracy of file access predictions and reduce loss due to incorrect prefetches. Each expert consists of files that have previously succeeded the current file along with one expert that predicts no successor is the best choice. By careful updating of the expert weights, file predictions can be made very accurately both on a p...

In this study we present a Deep Mixture of Experts (DMoE) neural-network architecture for single microphone speech enhancement. By contrast to most speech enhancement algorithms that overlook the speech variability mainly caused by phoneme structure, our framework comprises a set of deep neural networks (DNNs), each one of which is an ‘expert’ in enhancing a given speech type corresponding to a...

This chapter presents a thorough review of current practice in eliciting, representing and amalgamating knowledge from multiple experts, with a focus on the use of graphical representations to support the process. The important points of the discussion are illustrated by presenting a specific methodology for eliciting and combining knowledge from multiple experts. This methodology provides a st...

Face alignment, which is the task of finding the locations of a set of facial landmark points in an image of a face, is useful in widespread application areas. Face alignment is particularly challenging when there are large variations in pose (in-plane and out-of-plane rotations) and facial expression. To address this issue, we propose a cascade in which each stage consists of a mixture of regr...

With the increasing availability of data from various domains such as health care, finance, social networks, etc. there is a need to provide analytic tools that are more accessible to lay people. In this paper, we present a software tool which can be used to aid inexperienced users in understanding high dimensional data. To facilitate the understanding of such data, we place special emphasis on...

In on-line machine learning, predicting changes is not a trivial task. In this paper, a novel prediction approach is presented, that relies on a committee of experts. Each expert is trained on a specific history of changes and tries to predict future changes. The experts are constantly modified based on their performance and the committee as a whole is thus dynamic and can adapt to a large vari...

Many problems, like feature selection, involve evaluating objects while ignoring the relevant underlying properties that determine their true value. Generally, an heuristic evaluating device (e.g. filter, wrapper, etc) is then used with no guarantee on the result. We show in this paper how a set of experts (evaluation function of the objects), not even necessarily weakly positively correlated w...

In this fast developing technical era of digital technology, there is an explosion of video data especially on the internet. Organizing video and locating required information effectively and accurately presents a great challenge to the researchers. This demands a tool which would break down the video into smaller and manageable units called shots. A wide range of approaches have been investiga...

This unique two-week elective, offered each February, is designed for fourth year medical students and residents in Internal Medicine, Family Medicine, and other disciplines who have an interest in public health and health policy but no advanced degree in these subjects (MPH or JD). The course combines class lectures by health policy experts and site visits in the Washington, D.C. area with a g...