| Parisi D. (1997). An Artificial Life approach to language. Mind and Language, 59, 121-146. |
....types of communication systems. Some rely on the use of simple signals, while others use symbolic communication systems or complex syntactical structures. Amongst the different types of computational approaches, evolutionary computation techniques, such as the synthetic approach of artificial life [20, 25], can be used to study the emergence of communication. This approach permits the study of the different stages of semiotic complexity, from simple associations between signals and objects to symbolic representations, and then to complex syntactic relationships between words. Evolutionary ....
Parisi D. (1997). An Artificial Life approach to language. Mind and Language, 59, 121-146.
....and plasticity with regard to the configuration of those components in relation to one another. Cellular Automata as a Generic Class of Embodiable Dynamical System. In contrast to the specific models involved in dynamical software techniques such as Artificial Life Neural Networks (ALNNs) [27] and some evolutionary models [28, 29] that recognise the significance of system environment interaction, it is proposed that Cellular Automata (CA) are appropriate for capturing the structure and inherent dynamics of embodied systems, of whatever form or ontological status (cf. 30] Boolean ....
Parisi, D.: An Artificial Life approach to language. Brain & Cog. 59 (1997) 121-146
....and Menczer s LEE (1996) Holland s ECHO (1995) and many others) is the idea of a population of individuals that interact in their world. This is extremely important when studying a behavior such as communication. Communication does not happen in a vacuum; it is a population level phenomenon (Parisi 1997). AL simulations of a population of communicators demonstrate how important the environment and other communicators are to the study of communication. 1 See Hendriks Jansen (1998) for a deeper discussion of these issues. Other models of the evolution of communication and language focus on form ....
Parisi, Domenico. 1997. An artificial life approach to language. Brain and Language, 59, pp. 121--146.
....protocol involves small groups of agents (2 or 3) that exchange signals about a source of food with the evolved continuous signals functioning as modulators of the agents behavior. For a general discussion of how to study language with neural networks in an Artificial Life perspective, cf. Parisi, 1997. In the present paper we describe some simulations on the evolutionary emergence of a very limited language , made up of just two one word utterances, in a population of simple organisms living in a simple environment. The behavior of each organism is controlled by a neural network and the ....
.... symbolic representations , while the hidden unit activations correspond to categorical representations since their patterns tend to maximize inter categorical differences and minimize intra categorical differences. For a discussion of these differences cf. Sharkey and Jackson, 1994; cf. also Parisi, Denaro, and Cangelosi, 1997. A further step in the evolution of a truly human language, of course, would be the evolution of an ability in neural networks to combine signals (and the internal categorical representations associated with them) to form complex signals with a syntax. If this can be ....
Parisi, D. (1997). An Artificial Life approach to language. Brain and Language.
....Networks (ALNN) are neural networks controlling the behavior of organisms that live in an environment and are members of evolving populations of organisms. ALNN models have been used to simulate the evolution of language (Cangelosi Parisi, 1998; Cangelosi, 1999; Cangelosi Harnad, in press; Parisi, 1997). For example, in Cangelosi and Parisi s (1998) model organisms evolve a shared lexicon for naming different types of foods. Communication signals are processed by neural networks with genetically inherited connection weights and the signals evolve at the population level using a genetic algorithm ....
....a genetic algorithm with no changes during an individual s lifetime. ALNN models provide a unifying methodological and theoretical framework for cognitive modeling because of the use of both evolutionary and connectionist techniques and the interaction of the organisms with a simulated ecology (Parisi, 1997). All behavioral abilities (e.g. sensorimotor skills, perception, categorization, language) are controlled by the same neural network. This unified framework permits the study of various factors affecting language evolution, such as the differences between genetic and learned communication ....
Parisi, D. (1997). An Artificial Life approach to language. Mind and Language, 59, 121-146.
No context found.
Domenico Parisi. An artificial life approach to language. Brain and Language, in press.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC