The paper presents a concept, implementation and real examples of dynamic parallelization of computations using services derived from MPI applications deployed in the BeesyCluster ...
Abstract. Although evolutionary algorithms (EAs) are often successfully used for the optimization of dynamically changing objective function, there are only very few theoretical re...
This paper presents a new workload model, called the state-dependent deadline model, for applications whose high-level timing requirements may change with time. The problem is how...
In this paper, we propose a new conceptual method for the design, investigation, and evaluation of multi-objective variation operators for evolutionary multi-objective algorithms. ...
A successful interpretation of data goes through discovering crucial relationships between variables. Such a task can be accomplished by a Bayesian network. The dark side is that, ...