We propose an algorithm to construct classification models with a mixture of kernels from labeled and unlabeled data. The derived classifier is a mixture of models, each based o...
We give an efficiently executable specification of the global constraint of lexicographic order in the Constraint Handling Rules (CHR) language. In contrast to previous approaches...
We have developed a new GA hardware called GAA-I (Genetic Algorithm Accelerator-I), in which the crossover operator to be applied to each individual was dynamically selected during...
The construction process is subject to an array of influences, both from internal and external environments, which makes the process performance uncertain and difficult to predict...
We investigate the problem of eliciting CP-nets in the well-known model of exact learning with equivalence and membership queries. The goal is to identify a preference ordering wi...