Our publications
[Novoa2009] Controlling Particle Trajectories in a Multi-swarm Approach for Dynamic Optimization Problems,
, International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2009, 2009, Volume 5601, Santiago de Compostela, Spain, p.285-294, (2009)
Abstract
In recent years, particle swarm optimization has emerged as a suitable optimization technique for dynamic environments, mainly its multi-swarm variant. However, in the search for good solutions some particles may produce transitions between non improving ones. Although this fact is usual in stochastic algorithms like PSO, when the problem at hand is dynamic in some sense one can consider that those particles are wasting resources (evaluations, time, etc). To overcome this problem, a novel operator for controlling particle trajectories is introduced into a multi-swarm PSO algorithm. Experimental studies over a benchmark problem shows the benefits of the proposal.
None
None
[Pelta2009] A study on diversity and cooperation in a multiagent strategy for dynamic optimization problems,
, International Journal of Intelligent Systems, 04/2009, Volume 24, Issue 7, p.844-861, (2009)
Abstract
In recent years, biological and natural processes have been increasingly influencing methodologies in science and technology. In particular, the role played by the cooperation among individuals is being studied more frequently and profoundly in diverse areas of knowledge.
We present here a multiagent decentralized strategy for dynamic optimization problems where a population of cooperative agents and solutions are used to deal with the moving peaks problem. We focus on cooperation and diversity mechanisms, and we study how different alternatives affect the performance of the strategy.
None
None
[Pelta2009b] Simple control rules in a cooperative system for dynamic optimisation problems,
, International Journal of General Systems, 10/2009, Volume 38, Number 7, p.701-717, (2009)
Abstract
The use of centralised, multi-threads cooperative systems, has emerged as a successful alternative to deal with static optimisation problems, avoiding the problem of selecting a particular, isolated strategy. However, when the problem to deal with becomes dynamic in some sense, the question as to whether those systems and the type of control rules employed to control the threads are useful or not remains open.
In this article, we depart from a strategy that joint uses a set of solutions and a set of simple agents, and we propose, compare and test two control rules for updating the former. The rules are a simple replacement frequency mechanism and a fuzzy set based one.
Computational experiments are performed on the moving peaks benchmark problem under different scenarios and the main conclusions are: first, the fuzzy set based rule is better than the frequency based rule and second, both rules are competitive when compared with a state-of-the-art algorithm.
None
