A novel optimisation technique inspired by natural ecosystems is presented, where the optimisation works at two levels: a first optimisation, migration of services which are distributed in a decentralised peer-to-peer network, operating continuously in time; this process feeds a second optimisation based on evolutionary computing that operates locally on single peers and is aimed at finding solutions to satisfy locally relevant constraints. Through this two-fold process, the local search is accelerated and will yield better local optima, because the distributed optimisation already provides a good sampling of the search space by making use of computations already performed in other peers with similar constraints. We call this new distributed optimisation architecture a digital ecosystem, an ecosystem-orientated architecture (EOA) created by extending a service-oriented architecture (SOA) with distributed evolutionary computing (DEC). The digital ecosystem will allow services to recombine and evolve over time, constantly seeking to improve their effectiveness for the user base. individuals within our digital ecosystem will be applications (groups of services), created in response to user requests by using evolutionary optimisation to aggregate the services. These individuals will migrate through the digital ecosystem and adapt to find niches where they are useful in fulfilling other user requests
5 March 2007 in IEEE Transactions on Neural Networks
In this letter, we comment on "Pruning Error Minimization in Least Squares Support Vector Machines" by B. J. de Kruif and T. J. A. de Vries. The original paper proposes a way of pruning training examples for least squares support vector machines (LS SVM) using no regularization (gamma = infin). This causes a problem as the derivation involves inverting a matrix that is often singular. We discuss a modification of this algorithm that prunes with regularization (gamma finite and nonzero) and is also computationally more efficient.
18 August 2010 in International Journal of Computer Vision