W.B.Langdon . 3 Jan 2017 2016 papers , full list
W. B. Langdon Technical Report RN/16/10. BIORXIV/2016/095075 DOI:10.1101/095075
ABSTRACT
Typically BarraCUDA
uses CUDA graphics cards to map DNA
reads to the human genome. Previously its software source code was
genetically improved for short paired end next generation sequences
[Langdon:2015:GECCO].
On longer, 150 base paired end noisy
Cambridge Epigenetix's
data,
a Pascal GTX 1080 processes about 10000 strings per second, comparable
with twin nVidia Tesla K40.
William B. Langdon and Justyna Petke and Bobby R. Bruce, Technical Report RN/16/01
ABSTRACT
A simple model of distributed Genetic Improvement running in parallel across a local area network in which start/stop commands are sent to measuring devices calculates a minimum usable software mutation effect based on the analogue to digital convert (ADC)'s resolution. With modern low cost power monitors, the high speed Ethernet LAN's jitter and delays appear to have little effect. Where the software to be improved permits it, optimal test duration is inversely proportionate to minimum mutation effect size, typically well under a second.
William B. Langdon and Javier Dolado and Federica Sarro and Mark Harman, Information and Software Technology, 73 (May 2016) pages 16-18. DOI:10.1016/j.infsof.2016.01.003 PDF
ABSTRACT
Shepperd and MacDonell ``Evaluating prediction systems in software project estimation''. Information and Software Technology 54 (8), 820-827, 2012, proposed an improved measure of the effectiveness of predictors based on comparing them with random guessing. They suggest estimating the performance of random guessing using a Monte Carlo scheme which unfortunately excludes some correct guesses. This biases their MARP0 to be slightly too big, which in turn causes their standardised accuracy measure SA to over estimate slightly. In commonly used software engineering datasets it is practical to calculate an unbiased MARP0 exactly.