ECTA (2) – Swarm Art with KANTS

The second paper we have presented last weekend in Barcelona (ECTA 2012), with the title “Pherogenic Drawings” (C.M. Fernandes, A.M. Mora, J.J. Merelo, A.C.Rosa), is about swarm art and describes the application of the KANTS algorithm, an ant-clustering algorithm created by our group in 2008, to data extracted from sleep electroencephalogram (EEG) signals.

We execute KANTS on a set of Hjorth parameters extracted from the EEG of sane adult humans. In the end of the run, we translate the environmental vectors of the algorithm into an RGB image. The resulting drawings are different for each patient, since the ants, in this algorithm, are the data samples, which communicate via the environmental grid of vectors and change that same environment. Therefore, a pherogenic (from pheromone+genesis) drawing is a signature of a person’s night sleep.

The results are contextualized within the swarm and generative art — a contemporary trend that blends art, science, technology — and within my own work on generative art (check the project that won the last GECCO’s competition on Evolutionary Art, Design and Creativity, also created with KANTS). The abstract:

Social insects and stigmergy have been inspiring several significant artworks and artistic concepts that question the borders and nature of creativity. Such artworks, which are usually based on emergent properties of autonomous systems and go beyond a centralized human authorship, are a part of a contemporary trend known as generative art. This paper addresses generative art and presents a set of images generated by an ant-based clustering algorithm that uses data samples as artificial ants. These ants interact via the environment and generate abstract paintings. The algorithm, called KANTS, consists of a simple set of equations that model the local behavior of the ants (data samples) in a way that, when travelling on a heterogeneous 2-dimensional lattice of vectors, they tend to form clusters according to the class of each sample. The algorithm was previously proposed for clustering and classification. In this paper, KANTS is used outside a purely scientific framework and it is applied to data extracted from sleep-Electroencephalogram (EEG) signals. With such data sets, the lattice vectors have three variables, which are used for generating the RGB values of a colored image. Therefore, from the actions of the swarm on the environment, we get 2-dimensional colored abstract sketches of human sleep. We call these images pherogenic drawings, since the data used for creating the images are actually the pheromone maps of the ant algorithm. As a creative tool, the method is contextualized within the swarm art field.

Ant the presentation:

ECTA 2012 (1) – Self-Organized Criticality and PSO (best paper award)

Last weekend we were in Barcelona presenting two papers at the 4th International Conference on Evolutionary Computation Theory and Applications (ECTA 2012). One is an extension of the work presented at PPSN in the last September and its title is “Using Self-Organized Criticality for Adjusting a Particle Swarm” (C.M. Fernandes, J.J. Merelo and A.C. Rosa). We use the Bak-Sneppen model (which is known to display Self-Organized Critically) for controlling the parameters of the Particle Swarm Optimization (PSO) algorithm. We test the proposed strategy on two different topologies for the swarm and show that the performance is very stable throughout the test set. The paper and the corresponding talk won the best paper award of the congress. This is the abstract:

The local and global behavior of Self-Organized Criticality (SOC) systems may be an efficient source for controlling the parameters of a Particle Swarm Optimization (PSO) without hand-tuning. This paper proposes a strategy based on the SOC Bak-Sneppen model of co-evolution for adjusting the inertia weight and the acceleration coefficients values of the PSO. In order to increase exploration, the model is also used to perturb the position of the particles. The resulting algorithm is named Bak-Sneppen PSO (BS-PSO). An experimental setup compares the new algorithm with versions of the PSO with varying inertia weight, including a state-of-the-art algorithm with dynamic variation of the weight value and perturbation of the particles’ positions. The parameter values generated by the model are investigated in order to understand the dynamics of the algorithm and explain its performance.

And this is the presentation:

Genebot (again) in CIG2012

Adaptative bots for real-time strategy game via map characterization(A.Fernández-Ares, P.García-Sánchez, A.M. Mora, J.J Merelo) is the title of the paper we have presented in CIG2012. In this work we use Genetics Algorithms for improve an adaptative bot for play (and win!) to planet wars. We made it through the characterization of the maps, studing those features (calculated quickly) that influence in bot behavior:  

The abstract:

This paper presents a proposal for a fast on-line map analysis for the RTS game Planet Wars in order to define specialized strategies for an autonomous bot. This analysis is used to tackle two constraints of the game, as featured in the Google AI Challenge 2010: the players cannot store any information from turn to turn, and there is a limited action time of just one second.They imply that the bot must analyze the game map quickly, to adapt its strategy during the game. Based in our previous work, in this paper we have evolved bots for different types of maps. 

Then, all bots are combined in one, to choose the evolved strategy depending on the geographical configuration of the game in each
Several experiments have been conducted to test the new approach, which outperforms our previous version, based on an off-line general training.