Derivations of variational gaussian process approximation framework

Recently, within the VISDEM project (EPSRC funded EP/C005848/1), a novel variational approximation framework has been developed for inference in partially observed, continuous space-time, diffusion processes. In this technical report all the derivations of the variational framework, from the initial work, are provided in detail to help the reader better understand the framework and its assumptions.

A genetic algorithm to optimise the component placement process in the PCB assembly

In printed circuit board (PCB) assembly, the efficiency of the component placement process is dependent on two interrelated issues: the sequence of component placement, that is, the component sequencing problem, and the assignment of component types to feeders of the placement machine, that is, the feeder arrangement problem. In cases where some components with the same type are assigned to more than one feeder, the component retrieval problem should also be considered. Due to their inseparable relationship, a hybrid genetic algorithm is adopted to solve these three problems simultaneously for a type of PCB placement machines called the sequential pick-and-place (PAP) machine in this paper. The objective is to minimise the total distance travelled by the placement head for assembling all components on a PCB. Besides, the algorithm is compared with the methods proposed by other researchers in order to examine its effectiveness and efficiency.

Research with in-built replications:comment and further suggestions for replication research

This brief commentary on the paper Designing Research with In-built Differentiated Replication expands on concerns about a lack of replication research by focusing on three key questions of continuous importance: Why should researchers conduct more replication research? Why do so few researchers conduct replication studies? What can researchers do to add more studies? This paper identifies barriers preventing replication related to the scientific system, the replication researcher, and the initial research. Suggestions include publishing all papers electronically along with reviews; authors taking steps to encourage replications of their work; and editors inviting replications of important papers. The scientific community should establish a replication index as a measure of output quality. © 2012 Elsevier Inc.