We describe a Markov chain Monte Carlo (MCMC)-based algorithm for sampling solutions to mixed Boolean/integer constraint problems. The focus of this work differs in two points from...
The process of extracting useful knowledge from large datasets has become one of the most pressing problems in today’s society. The problem spans entire sectors, from scientists...
: Many classification problems involve high dimensional inputs and a large number of classes. Multiclassifier fusion approaches to such difficult problems typically centre around s...
We describe the use of the discrete wavelet transform (DWT) for non-parametric linear time-invariant system identification. Identification is achieved by using a test excitation t...
Object-oriented legacy systems are hard to maintain because they are hard to understand. One of the main understanding problems is revealed by the so-called ”yo-yo effect” tha...