We present a model-free reinforcement learning method for partially observable Markov decision problems. Our method estimates a likelihood gradient by sampling directly in paramet...
In this paper, we are interested in the relationship between final prices of online auctions and possible shill activities during those auctions. We conduct experiments on real auc...
Different algorithms have been proposed in the literature to cluster gene expression data, however there is no single algorithm that can be considered the best one independently on...
We propose a real-time pose invariant face recognition algorithm from a gallery of frontal images only. First, we modified the second order minimization method for active appearan...
Abstract. In this paper, we test some of the most commonly used classifiers to identify which ones are the most robust to changing environments. The environment may change over tim...
Houman Abbasian, Chris Drummond, Nathalie Japkowic...