[Show abstract][Hide abstract] ABSTRACT: We investigated the minimal condition for symmetry breaking in morphogenesis of cellular population using cellular automata based on reaction-diffusion dynamics. We started to understand the morphogenic process and provide the mathematical formulations of the fields aassociated with cellular activity, environmental struture and dynamics, and mutual influence of the environment in cells activity ans self-organisation. Then we model morphogenesis in cellular clusters, starting from a single seed cell, in the context of some environment, considering a variety of cellular and environmental processes, and the interaction between the two types of processes in the cellular dynamics. In particular, we looked for the possibility of the emergence of branching structures due to mechanical interactions. The model used two types of cells an external gradient. The results showed that the external gradient influenced movement of cell type-I, also revealed that clusters formed by cells type-II worked as barrier to movement of cells type-I.
[Show abstract][Hide abstract] ABSTRACT: We present a model of a recurrent neural network, embod- ied in a minimalist articulated agent with a single link and joint. The configuration of the agent defined by one angle (degree of freedom), is determined by the activation state of the neural network. This is done by contracting a muscle with many muscular fibers, whose contraction state needs to be coordinated to generate high amplitude link displacements. In networks without homeostasic (self-regulatory) mechanism the neural state dynamics and the configuration state dynamics converges to a fixed point. Introduction of random noise, shows that fixed points are meta- stable. When neural units are endowed with homeostasic mechanisms in the form of threshold adjustment, the dynamics of the configuration angle and neural state becomes aperiodic. Learning mechanisms foster functional and structural cluster formation, and modifies the distribution of the kinetic energy of the network. We also present a meta-model of embodied neural agents, that identifies self-perturbation as a mechanism for neural development without a teacher.
Artificial Neural Networks - ICANN 2007, 17th International Conference, Porto, Portugal, September 9-13, 2007, Proceedings, Part II; 01/2007
[Show abstract][Hide abstract] ABSTRACT: We present a model of a recurrent neural network with home- ostasic units, embodied in a minimalist articulated agent with a single link and joint. The configuration of the agent is determined by the total activation level or kinetic energy of the network. We study the complex- ity patterns of the neural networks, and see how the entropy of the neural controller state and agent configuration changes with the relative char- acteristic time of the homeostasis when compared with the excitatory- inhibitory activation dynamics of network. We also present a meta-model of embodied neural agents, that serves as conceptual framework to study self-perturbation and the self-organization in embodied neural agents. Simulation results show that homeostasis significantly influences the dy- namics of the network and the controlled agent, allowing the system to escape fixed-points and produce complex aperiodic behavior. The rela- tion between the characteristic time of homeostasis and the characteristic time of main excitatory-inhibitory activation dynamics was found to be non-linear and non-monotonic. We use these findings to connect the per- spectives of classical cybernetics on homeostasis to complexity research.
Advances in Artificial Life, 9th European Conference, ECAL 2007, Lisbon, Portugal, September 10-14, 2007, Proceedings; 01/2007
[Show abstract][Hide abstract] ABSTRACT: We present a simple model of self-organization for sensor networks that addresses two conflicting requirements: hazard situations should be reported to a sink node with as little delay as possible, even in highly dynamic regimes; and power consumption by individual nodes should be as low as possible and balanced. The model includes a surveillance protocol that explores correlation between source location and event types, and a variation of gradient-based routing that adapts continuously to energy available at selected routers and to changes in topology. Simulation runs provide support to the heuristics we implemented to select routers and nodes to report events, since network longevity increases when compared to other solutions for sensor networks. The performance increase is particularly accentuated when the correlation between event types at neighboring nodes is significant.
Engineering Self-Organising Systems, 4th International Workshop, ESOA 2006, Hakodate, Japan, May 9, 2006, Revised and Invited Papers; 01/2006
[Show abstract][Hide abstract] ABSTRACT: The distribution of age at first marriage shows well-known strong regularities across many countries and recent historical periods. We accounted for these patterns by developing agent-based models that simulate the aggregate behavior of individuals who are searching for marriage partners. Past models assumed fully rational agents with complete knowledge of the marriage market; our simulated agents used psychologically plausible simple heuristic mate search rules that adjust aspiration levels on the basis of a sequence of encounters with potential partners. Substantial individual variation must be included in the models to account for the demographically observed age-at-marriage patterns.
[Show abstract][Hide abstract] ABSTRACT: We present a framework for a Multi-Agent System (MAS) devised to support the modelling and simulation of agent-based models of human social behavior and culture change. We set forth its main abstractions, and test the usefulness and generality of the framework by describing how two previously published models from the literature have been re-implemented in it. We argue that our framework provides features that simplify the modelling process of a wide range of models of human social behavior, beyond what current MAS accomplish.
[Show abstract][Hide abstract] ABSTRACT: We present a model of human mate choice that shows how realistic population-level patterns of assortative mating can self-organize and emerge from the behavior of individuals using simple mate search rules. In particular, we model plausible psychological mechanisms for mate search and choice in a realistic social ecology. Through individual interactions, patterns emerge that match those observed in typical human societies, particularly with regard to correlated quality levels within couples, distributions of the ages at which couples mate, and effects of skewed sex ratios on these mating age distributions.
Artificial Life 02/2003; 9(4):403-17. DOI:10.1162/106454603322694843 · 1.93 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: I claim that the increase in complexity in the (known) trace of Paleolithic stone tools can be parsimoniously explained by postulating the emergence of effective mechanisms for the social transmission of representations. I propose that Paleolithic tools, similar to more contemporary tools, were subject to a process of evolution by artificial selection based on functionality.
[Show abstract][Hide abstract] ABSTRACT: We present a conceptual framework for the study of mate choice in monogamous mating systems with non-negligible courtship time. Within this framework, we develop a mate choice model for the common case where individuals have a changing social network of potential partners. The performance and robustness of different agent strategies is evaluated, emphasizing the important role that courtship plays in mate choice. Specifically, the courtship period can be used by individuals to swap to better partners when they become available. We found that using courtship as a mechanism for holding partners before full commitment to mating provides strategic advantages relative to sequential search using aspiration levels. Moreover, simple heuristics that require little computation provide a degree of robustness to environmental (parameter) changes that is unattainable by strategies based on more extensive information processing. Our model produces realistic patterns of assortative mating (high within-couple mate value correlations) and rates of mating that match empirical data on human sexual/romantic relationships much more closely than previous accounts from biology and the social sciences.
[Show abstract][Hide abstract] ABSTRACT: We present a new model of human mate choice incorporating non-negligible courtship time. The courtship period is used by individu- als to strategically swap to better partners when they become available. Our model relies on realistic assumptions about human psychological constraints and the specifics of the human social environment to make predictions about population level patterns that are supported by em- pirical data from the social sciences.
Advances in Artificial Life, 6th European Conference, ECAL 2001, Prague, Czech Republic, September 10-14, 2001, Proceedings; 01/2001
[Show abstract][Hide abstract] ABSTRACT: Human social behaviour, culture change, and emergent social organization are amongst themost intricate phenomena studied by science. Aided by theoretical and computational toolsdeveloped to study emergent phenomena in complex systems, social theorists aim to develop aunied body of knowledge that helps to shed light on long lasting question of human sociality.
[Show abstract][Hide abstract] ABSTRACT: We present a generic framework for embodied neural agents relying on the interplay between three types of self-organization mechanisms: learning, homeosta- sis, and self-perception. Agents are characterized at two levels: the micro or neural level, containing a high number of degrees of freedom, and the macro or configuration level, containing a much lower num- ber of degrees of freedom (e.g., the joint angles of an articulated agent). Learning and homeostasis in neu- ral units are understood as upward causality (emer- gence) mechanisms, and self-perception (the neural controller receiving feed-back from the eects of its own activity at the macro level) are understood as a downward causality mechanism. The theoretical analysis presented relates the evolution of the neural state space and changes in its probability distribution with the evolution of the configuration state space. The framework seeks to explains how complex (bio- )physical systems can develop adaptive behavior and cognitive skills with minimal external support (e.g., from a teacher). To illustrate the practical use of the framework, we present a model of a minimalist artic- ulated agent with a single link and joint, controlled by a recurrent neural network with homeostasic units, and use computer experiments to study the dynam- ics of the system looking at and relating the neural
[Show abstract][Hide abstract] ABSTRACT: We present a minimalist model of an embodied neural agent for visual attention tasks. The model is composed of a simple articulated struc- ture with a single link, a tip sensitive to visual stimuli, and an agnostic- antagonistic muscle pair. The body configuration of the agent is defined by the link angle (one degree of freedom), and depends on dierences in muscular force. The level of muscular contraction-distension is controlled by a recurrent neural network. The neural network consists of arrays of neural units fully connected, that receive input (from visual stimuli) from the environment. Fixed and moving point particles are used as sources of visual stimuli. The neural network includes two modes of operation — with and without an adaptive threshold (modeling units' homeosta- sis). The results show that visual accuracy with homeostasis is higher than without homeostasis. This happen because homeostasis increases the complexity of neural and agent configuration dynamics, and make less likely that the agent reaches a fixed point without detecting the stimulus. The agent also demonstrates the ability to adjust the visual alignment to new angular positions and abrupt changes, and to follow particle moving with a uniform velocity.