We present a new type of multi-class learning algorithm called a linear-max algorithm. Linearmax algorithms learn with a special type of attribute called a sub-expert. A sub-exper...
We introduce a method for learning Bayesian networks that handles the discretization of continuous variables as an integral part of the learning process. The main ingredient in th...
Our goal is to optimize regularized image reconstruction methods for emission tomography with respect to the task of detecting small lesions of unknown location in the reconstruct...
: Novelty detection, or anomaly detection, on temporal sequences has increasingly attracted attention from researchers in different areas. In this paper, we present a new framework...
It has been observed that even highly optimized software programs perform "redundant" computations during their execution, due to the nature (statistics) of the values a...