Cloud-based evolutionary algorithms: An algorithmic study

This is a new paper of the Cloud Storare Services series. It has been accepted for publication in Natural Computing with DOI 10.1007/s11047-012-9358-1. Inside you can find a study of how an evolutionary algorithm implemented using two different and free cloud storage services as Dropbox and SugarSync evolves a population of individuals for solving traditional problems like MMDP and P-PEAKS. The selected resources for this case have been homogeneous and heterogeneous computers for the same problems and the same experiments. Results underline SugarSync as the best option for these problems in a local network with homogeneous computers because the quality of the solutions is better using SugarSync than Dropbox although the differences are not significatives.

You can find the work with all the information on Springer .

Advertisements

Science and Videogames Tutorial

Last 16th November, inside the GAME-ON 2012 Conference, I presented (with Antonio Fernández Leiva) a tutorial entitled “Computational Intelligence applied to videogames; past, present and future”.

It was a two parts tutorial, being the first one (mine) devoted to introduce the relationship between science and videogames, describing the integration of Computational Intelligence in this scope.

My part presentation can be seen here:

Enjoy it! :D

Evolutionary Algorithms in Heterogeneous Nodes

Today I presented a brief talk about some papers about the usage of heterogeneous computers for distributed EAs:

All these ideas (and new ones) are being applied in our Service Oriented Architecture for Evolutionary Algorithms, we hope to show interesting results soon!

Here is the presentation (in Spanish).

ICCS’12 – Swarms, Complexity and Art

We have just back from the International Congress on Complex Systems, held in Agadir, Morocco, were we have presented the papers “Swarm Art with KANTS” and “Towards a 2-dimensional Self-organized Framework for Structured Population-based Metaheuristics”. The first one continues our line of work with the KANTS algorithm as a swarm art tool and describes drawings generated by data extracted from photographs.

Original photos and respective pherogenic drawings by the KANTS algorithm

The second paper, which opens a new line of research, describes a swarm system that, guided by simple rules and with no central coordination, is driven to a state in which global patterns emerge. In that state, the components of the swarm self-organize into highly dynamic clusters. We show that the system is unpredictable and robust. We also demonstrate that the system’s variables (averaged clustering degree and averaged distance between neighboring components) dysplay 1/f noise. The abstract:

This paper proposes a swarm intelligence framework for distributed population-based metaheuristics that uses stigmergy and similarity measures as basic modeling rules with a local range of action for structuring the neighborhood. The system ­– which can be described as a cellular automaton with short-term memory – displays complex and emergent behavior whose most visible trait is the self-organization of a population of particles into dynamic clusters. These clusters tend to gather similar particles (similarity here is measured as the algebraic difference between randomly assigned fitness values). During the execution of the algorithm, the particles move through a grid of nodes leaving a mark with the fitness value of the particle in each node they visit. When deciding where to move, the particles take into account the marks in the neighborhood and tend to travel to nodes with marks that minimize the difference between the particle’s fitness and the mark’s fitness. A kind of hierarchical behavior is also modeled by forcing the particles to move toward nodes with better fitness values. We show that these simple rules conduct the system to a critical state in which clusters are constantly created and broken, while maintaining a typical pattern of clusters and paths. In addition, we demonstrate that the system’s variables display  noise, which is one of the signatures of Self-Organized Criticality (SOC). Since it does not require the tuning of control parameters to precise values, we hypothesize that the proposed system converges to SOC.

The presentation: