Marco A. Ramos

An action selection process to simulate the human behavior in virtual humans with real personality

In this paper, we present an action selection mechanism to simulate the human behavior in realistic and believable virtual humans according to their personality and affective state taking into account their beliefs, desires, intentions, and the level of intensity of the events they perceive from their environment.

A Middleware for Integrating Cognitive Architectures

The features of distributed systems help to solve problems in different research areas like fault tolerance, use of distributed resources, etc. The relevant cognitive architectures (CA) use middleware (distributed systems concept) to test its models and propose new theories. Thanks to a middleware, the researchers may conceive CAs as a whole, not as a set of components. However, most of the middlewares used in present CAs are modifications of generic ones, which leads to extra processing affecting the whole performance. In this research, we propose a middleware designed and developed taking into account the requirements of CAs. Our middleware allows us the integration of different cognitive functions, like memory and attention developed independently in an easily and incrementally. Also our middleware allows us test the cognitive functions integrated in the CA. To test our proposal, the middleware simulates an attention-novelty handling cognitive process.

An emotional regulation model with memories for virtual agents

Emotional regulation is a mechanism to adjust our behavior to the current environment. We use this mechanism to achieve goals and objectives. This paper proposes a model for emotional regulation in virtual agents based on biological evidence of human brain function. The evidence shows different brain activation during emotional regulation. The model defines the techniques that are implemented, the data flow, and the data processing in each brain area during the emotional regulation. In the case study, a virtual agent shows behavior changes when it takes into account the emotional regulation mechanism and when it does not. The agent has emotional memories; they are from previous experiences and help to provide the desired behavior.

Semantic death in plant's simulation using Lindenmayer systems

Plants simulation through Lindenmayer Systems is a well know field, but most of the work in the area focus on the growth part of the developmental process. From an artificial life perspective, it is desired to have a simulation that includes all the stages of the cycle of life of a plant. That is the reason why this paper target the last stage and propose a strategy to include the concept of death through Lindenmayer systems. By using parametric and context-sensitive Lindenmayer systems in the modeling and simulation, the semantics of the mentioned concept can be captured and, thereby, with the proper interpretation, a graphic result, at a morphological level, can be displayed. A proof of concept that includes most of the concepts covered is also given.

A Fuzzy Model to Update the Affective State of Virtual Humans: An Approach Based on Personality

In this paper, we present a fuzzy mechanism to update in a more natural way the emotional and mood states of virtual humans. To implement this mechanism, we take into account the ten personality scales defined by Minnesota Multiphasic Personality Inventory to endow virtual humans with a real personality. In this manner, we apply different sets of fuzzy rules to change and regulate the affective state of virtual humans according to their personality, emotional and mood history, and the level of intensity of events they perceive from their environment.

A Computational Model of Emotional Attention for Autonomous Agents

A major challenge in artificial intelligence has been the development of autonomous agents (AAs) capable of displaying very believable behaviors. In order to achieve such objective, the underlying architectures of these intelligent systems have been designed to incorporate biologically inspired components. It is expected that through the interaction of this type of components, AAs are able to implement more intelligent and believable behavior. Although the literature reports several computational models of attention and emotions developed to be included in cognitive agent architectures, these have been implemented as two separated processes, disregarding essential interactions between these two human functions whose modeling and computational implementation may increase the believability of behaviors developed by AAs. In this paper, we propose a biologically inspired computational model of emotional attention. This model is designed to provide AAs with adequate mechanisms to attend and react to emotionally salient elements in the environment. The results of four types of simulations performed to evaluate the behavior of AAs implementing the proposed model are presented.

Emotional Attention in Autonomous Agents: A Biologically Inspired Model

Attention and emotions are two major functions underlying human behavior. Evidence shows that these two processes interact extensively in the human brain. In fields such as human-computer interactions and artificial intelligence, computational models of attention and emotions have been developed to be included in cognitive agent architectures. However, these have been implemented as two separated processes. Although this strategy has allowed the development of intelligent agents capable of showing very believable behavior, the modeling of the interactions between the attention and emotion processes has been widely ignored. In this paper, we propose a biologically inspired computational model of emotional attention. This model is designed to provide intelligent agents with adequate mechanisms to attend and react to emotionally salient elements in the environment. The simulations demonstrate that the proposed model helps increase the believability of virtual agents behaviors.

Chaotic Time Series Prediction with Feature Selection Evolution

Chaotic time series have been successfully predicted with the EPNet algorithm through the evolution of artificial neural networks. However, the input feature selection problem has either not been fully explored before or has not been compared against other algorithms in the literature. This paper presents four algorithms derived from the classical EPNet algorithm to evolve the input feature selection in three different chaotic series: Logistic, Lorenz and Mackey-Glass. Additionally, some flaws in the prediction field that may be considered in future works are discussed. A comparison against previous work demonstrates that in most cases the specialization of the EPNet algorithm allows better solutions with a smaller number of generations.

Autonomous agents and anticipative systems

In this paper we propose an artificial intelligence approach to real time simulation of a virtual community based on a multiagent system. Our approach is based on the work of Holland and Miller (1991), in which the authors propose that the system may be viewed as a complex dynamic adaptive system with a large number of different kinds of agents. In such environment, agents may take their decisions based on the anticipation of the future state of the world (Axelrod, 1997). In this paper, we explore how useful the concept of anticipation can be for real time simulation improving and facilitating decisions of the virtual entities. In this paper, we work with real time anticipative agents and where developed situations, actions, and changes in the world. This normal context in a virtual world is taken as a framework where an agent is able to compare previous learned situations, executed actions and obtained results; with current situations in order to decide which action could lead the agent to a situation with the best utility or satisfaction degree.

E-health: agent-based models to simulate behavior of individuals during an epidemic outbreak

Infectious diseases are a threat to human population. Governments around the world invest a lot of money on research of health area. Artificial intelligence techniques are useful to simulate real scenarios and save in public spending. In this paper, the authors present an agent based model of behavior and activities of individuals, according to a set schedule in a population within an urban environment, which is useful for innovation labs. The authors report results on simulations of the AH1N1 influenza epidemic outbreak of 2009 in Toluca, México. Conclusions indicate that the population density implies that a higher concentration of people corresponds to a higher probability of contagion. This parameter is influenced by the activities and interactions that the agents have within the simulation. The proposed model allows a broader perspective to the analysis of infectious disease in a population describing the behavior and interactions among individuals.