Deep Belief Networks (DBN's) are generative models that contain many layers of hidden variables. Efficient greedy algorithms for learning and approximate inference have allow...
We present a general framework for image discrimination based on identifying small, localized differences between images. Our novel matching scheme is based on an alternate inform...
Huzefa Neemuchwala, Alfred O. Hero, Paul L. Carson...
The field of algorithms for pairwisc biosequence similarity search is dominated by heuristic methods of high efficiency but uncertain sensitivity. One reason that more formal stri...
Abstract-- Sensor networking technologies have developed very rapidly in the last ten years. In many situations, high quality multimedia streams may be required for providing detai...
In this paper we present a statistical learning scheme for image classification based on a mixture of old fashioned ideas and state of the art learning tools. We represent input i...