Publications (70) View all
-
Article: The effects of the trophic level on the stability of food webs
[show abstract] [hide abstract]
ABSTRACT: The study of food webs has long been a central topic of ecological research, but the structural effects of trophic level on their stability are still not clear. The work described here addresses the influence of a restriction arising from the trophic level on the network topology of food webs which affects their global behavior. We propose a network model of food webs in which the degree of the effects of the trophic level on speciation can be adjusted continuously by a single parameter. The restriction limits the number of species at each level and the establishment of prey-predator relationships between distant levels. Experimental results show that the restriction contributes to the stability of the ecosystem. This is because the strong restriction keeps less robust species at the lower levels abundant by making the distribution of the number of species at each level flat, while the distribution became an inverse pyramidal structure without the restriction. On the other hand, we found that several features of the network, such as the power-law distribution of coextinction sizes and the number of predators, do not depend on the degree of restriction. We also show several comparisons of the experimental data with empirical data in the fossil records.Artificial Life and Robotics 04/2012; 14(3):379-383. -
Article: Autonomous acquisition of cooperative behavior based on a theory of mind using parallel genetic network programming
[show abstract] [hide abstract]
ABSTRACT: Understanding others as having intentional states, such as beliefs and desires, is called the theory of mind (ToM). To clarify the mechanism of the autonomous acquisition of cooperative behavior based on the ToM, we constructed a functional model of the brain based on the functional parts combination (FPC) model. This model consists of a set of functional parts and activation signals specifying selective activated patterns, and activated modules can be executed in parallel based on the flow of control tokens. The module network and activation signals can be acquired by the evolutionary computation techniques used in genetic network programming and the genetic algorithm, respectively. We used a hunter task as the task to be solved by agents, and encoded inherent activation signals into the genome as a first step. The result of a computer simulation shows the emergence of the pattern of the functional parts for processing the ToM through evolution characterized by punctuated equilibrium. Key wordsFunctional model of the brain–Theory of mind–Cooperation–Evolution–Genetic network programmingArtificial Life and Robotics 04/2012; 16(2):157-161. -
Chapter: Adaptive Walk on Fitness Soundscape
[show abstract] [hide abstract]
ABSTRACT: We propose a new IEC for musical works based on an adaptive walk on a fitness landscape of sounds. In this system, there is a virtual plane that represents the genetic space of possible musical works called fitness soundscape. The user stands on the soundscape, and hears the multiple sounds that correspond to one’s neighboring genotypes at the same time. These sounds come from different directions that correspond to the locations of their genotypes on the soundscape. By using the human abilities for localization and selective listening of sounds, the user can repeatedly walk toward the direction from which more favorite sounds come. This virtual environment can be realized by a home theater system with multiple speakers creating “surrounded sound”. We report on the basic concept of the system, a simple prototype for musical composition with several functional features for improvement of evolutionary search, and preliminary evaluations of the system. Keywordsinteractive evolutionary computation–musical composition–fitness landscape–soundscape–virtual reality–artificial life05/2011: pages 94-101; -
Article: A Constructive Approach to the Evolution of the Planning Ability.
IJALR. 01/2011; 2:22-35. -
Conference Proceeding: DE/isolated/1: A New Mutation Operator for Multimodal Optimization with Differential Evolution.
AI 2011: Advances in Artificial Intelligence - 24th Australasian Joint Conference, Perth, Australia, December 5-8, 2011. Proceedings; 01/2011
