Abstract. Spatially structured population models improve the performance of genetic algorithms by assisting the selection scheme in maintaining diversity. A significant concern wi...
Abstract— We develop a general framework for MAP estimation in discrete and Gaussian graphical models using Lagrangian relaxation techniques. The key idea is to reformulate an in...
Jason K. Johnson, Dmitry M. Malioutov, Alan S. Wil...
We address the challenge of semantic gap reduction for image retrieval through an improved SVM-based active relevance feedback framework, together with a hybrid visual and concept...
Recognition using appearance features is confounded by
phenomena that cause images of the same object to look different,
or images of different objects to look the same. This
ma...
Ali Farhadi, Mostafa Kamali Tabrizi, Ian Endres, D...
Abstract— The success of classical high level synthesis has been limited by the complexity of the applications it can handle, typically not large enough to necessitate the depart...