Blanca Cases

Further results on swarms solving graph coloring

We have proposed the mapping of graph coloring problems into swarm dynamics. Empirical evidence that flock steering behaviors augmented with the notion of hostility (enmity and friendliness) are enough to perform efficiently the task of coloring the nodes of graphs even in the case 3-coloration hard graph topologies. We discuss here what are the minimal cognitive capabilities that allow the emergent behavior of swarms to solve such NP-complete problem without mediating an explicit knowledge representation.

Topos 2: spiking neural networks for bipedal walking in humanoid robots

This work analyses the state of the art in the field of Evolutionary Robotics and marks the path we select in the design of evolutionary strategies. The aim of this text is to describe the lines that we are going to follow in the foreseeable future. Our goal is to create through evolution the neural network that couples with a complex humanoid robot body. For us the problems of a non-structured environment and of Evolutionary Robotics need a sub-symbolic conexionist approach based in Nouvelle AI that can cope with the coupling among sensorimotor, neural and environment parts. We also describe the tools we choose to accomplish this task.

Eco-Grammars to Model Biological Systems: Adding Probabilities to Agents

The aim of this paper is to define probabilistic Ecogrammar systems and some of their applications in the field of evolutionary biology. A probabilistic Eco-grammar system is composed of agents that select the rules of internal growth as well as the action rules according to distributions of probability. The environment is 0L probabilistic. Our interest is centered in the study of the normalized populations of symbols obtained along a derivation. In this paper we show that the probabilistic approach applied to Eco-grammar systems allows to model the evolutionary stable strategies of Maynard Smith.

Exemplar Selection Using Collaborative Neighbor Representation

Retrieving the most relevant exemplars in image databases has been a difficult task. Most of exemplar selection methods were proposed and developed to work with a specific classifier. Research in exemplar selection is targeting schemes that can benefit a wide range of classifiers. Recently, Sparse Modeling Representative Selection (SMRS) method has been proposed for selecting the most relevant instances. SMRS is based on data self-representation in the sense that it estimates a coding matrix using a codebook set to the data themselves. The matrix coefficients are estimated using block sparsity constraint. In this paper, we propose a coding scheme based on a two stage Collaborative Neighbor Representation in the matrix of coefficients is estimated without any explicit sparse coding. For the second stage, we introduce two schemes for sample pruning in the second stage. Experiments are conducted on summarizing two video movies. We also provide quantitative performance evaluation via classification on the selected prototypes. To this end, one face dataset, one handwritten digits dataset, and one object dataset are used. These experiments showed that the proposed method can outperform state-of-the art methods including the SMRS method.

On how percolation threshold affects PSO performance

Statistical evidence of the influence of neighborhood topology on the performance of particle swarm optimization (PSO) algorithms has been shown in many works. However, little has been done about the implications could have the percolation threshold in determining the topology of this neighborhood. This work addresses this problem for individuals that, like robots, are able to sense in a limited neighborhood around them. Based on the concept of percolation threshold, and more precisely, the disk percolation model in 2D, we show that better results are obtained for low values of radius, when individuals occasionally ask others their best visited positions, with the consequent decrease of computational complexity. On the other hand, since percolation threshold is a universal measure, it could have a great interest to compare the performance of different hybrid PSO algorithms.

Percolating swarm dynamics

Swarm Intelligence has been a successful approach to solve some combinatorial problems through the metaphor of interacting evolving individuals of a a population P in a closed torus-like space S. Each individual usually perceives an space sorrounding it which can be generally modelled as disk of radius R. In this paper we discuss that Percolation conditions can be key to allow convergence to reach the optimal results of these kind of systems.

Data reduction through kernel sparse coding

Sparse Modeling Representative Selection (SMRS) has been recently proposed for finding the most relevant instances in datasets. This method deploys a data self-representativeness coding in order to infer a coding matrix that is regularized with a row sparsity constraint. The method assumes that the score of any sample is set to the L2 norm of the corresponding row in the coding matrix. Since the SMRS method is linear, it cannot always provide good relevant instances. Moreover, many of its selected instances are already in dense areas in the input space. In this paper, we propose to alleviate the SMRS methods shortcomings. More precisely, We propose two kernel data self-representativeness coding schemes that are based on Hilbert space and column generation. Performance evaluation is carried out on reducing training image datasets used for recognition tasks. These experiments showed that the proposed kernel methods can provide better data reduction than state-of-the art selection methods including the SMRS method.

Chaotic patterns in crowd simulation

We have developed a model of crowd based on social agents and swarm intelligence. The model takes the form of a randomly generated directed graph where nodes represent individuals locked in a room where a fire occurs. Each individual follows connected individuals, called the references. There are two special individuals, the firefighters, situated at the two exits from the room: exit T or true and exit F or false. The agents are directed at targets T and F in a computer animation implemented in Netlogo. They come into conflict of over-information when they receive contradictory information from their references. We studied experimentally the influence of the following mechanisms of conflict resolution: follow the mode or the anti-mode, random resolution and the effect of excluding the own opinion. We found that these mechanisms lead people following the mode to unanimously choose one of the options, flocking towards the selected goal. In the case of anti-mode, the population oscillates between the two options. The number of references is critical to this behavior and following one or two references with exclusion of the own state leads the system to chaotic patterns of convergence.

Towards a proof of the decidability of the momentary stagnation of the growth function of DOL systems

This paper proves the decidability of several problems in the theory of HD0L, D0L and PD0L systems, some of which that have been proved before but are now proved in a different way. First, the paper tackles the decidability of the nilpotency of HD0L systems and the infinitude of PD0L languages. Then, we prove the decidability of the problem of momentary stagnation of the growth function of PD0L systems. Finally, we suggest a way to solve the decidability of the momentary stagnation of the growth function of D0L systems, proving the decidability of the infinitude of HD0L as a trivial consequence.

Modeling Chaotic Series by Simple Eco-Grammar Systems with Reproduction, Death and Maturation of Agents

Grammar systems are abstract models of computation, in the field of Formal Language Theory, created to study the agent systems proposed by Artificial Intelligence. Eco-Grammar or EG systems are grammatical models of ecosystems. Simple Eco-Grammar Systems or SEG systems are a sub-class of EG. In this work we propose a model of SEG systems whose agents have capabilities ofReproduction, Death and Maturation, called SEG[RDM]. They are able to represent the dynamics of the Logistic equation developed by May in 1973, which is one of the main references in the study of chaos in the dynamics of populations in ecosystems. This work is going on the direction of approaching Artificial Intelligence, Theoretical Computer Sciences and Complex Systems.