Systems, Actors and Agents: Operation in a multicomponent environment

📝 Abstract

Multi-agent approach has become popular in computer science and technology. However, the conventional models of multi-agent and multicomponent systems implicitly or explicitly assume existence of absolute time or even do not include time in the set of defining parameters. At the same time, it is proved theoretically and validated experimentally that there are different times and time scales in a variety of real systems - physical, chemical, biological, social, informational, etc. Thus, the goal of this work is construction of a multi-agent multicomponent system models with concurrency of processes and diversity of actions. To achieve this goal, a mathematical system actor model is elaborated and its properties are studied.

💡 Analysis

Multi-agent approach has become popular in computer science and technology. However, the conventional models of multi-agent and multicomponent systems implicitly or explicitly assume existence of absolute time or even do not include time in the set of defining parameters. At the same time, it is proved theoretically and validated experimentally that there are different times and time scales in a variety of real systems - physical, chemical, biological, social, informational, etc. Thus, the goal of this work is construction of a multi-agent multicomponent system models with concurrency of processes and diversity of actions. To achieve this goal, a mathematical system actor model is elaborated and its properties are studied.

📄 Content

1

Systems, Actors and Agents:
Operation in a multicomponent environment Mark Burgin
University of California, Los Angeles 405 Hilgard Ave. Los Angeles, CA 90095

Abstract. In this paper, we further develop multi-agent approach by creating new types of system models. The problem is that conventional models of multi-agent and multicomponent systems implicitly or explicitly assume existence of the absolute time or even do not include time in the set of defining parameters. However, it is rationalized theoretically and validated experimentally that there are different times and time scales in a variety of real systems – physical, chemical, biological, social, etc. Thus, the goal of this paper is construction of multi-agent and multicomponent system models with concurrency of processes and diversity of actions. To achieve this goal, a mathematically based system actor model is elaborated and its properties are studied. Keywords: time, system, actor, agent, action, process, interaction, environment

1. Introduction Multi-agent approach is becoming more and more popular in the area of computing, networking, artificial intelligence, robotics, distributed control, resource management, collaborative decision support systems, data mining, etc. (Buşoniu, et al, 2010; Shoham and Leyton-Brown, 2008; Vlassis, 2007; Weiss, 1999). Usually, it is assumed that a multi-agent system is a group of autonomous, interacting entities sharing a common environment, which they perceive with sensors and upon which they act with actuators. Time is a critically important characteristic of any real-life system. However, not all features of time are adequately presented in the conventional multi-agent models. If we analyze existing approaches and directions in the multi-agent approach, we can see that in all dynamic models of multi-agent systems, either time is implicitly induced by actions of agents and system states or it is explicitly assumed that unique time exists for the whole system. An archetypal 2

example of this situation is the absolute Newtonian time in the physical universe, which is innate for the entire classical physics.
However, relativity theory and various experiments disproved this assumption bringing forth the concept of local time (cf., for example, (Einstein, et al, 1923)). The system theory of time extends this principle much further (Burgin, 1992; 2002). Other researchers also advocated existence of different times or different time scales in their theories (cf., for example, (Prigogine, 1980; Barwise and Seligman, 1997)). Besides, as Norbert Wiener (1961) writes, one of the most famous philosophers of the 20th century Bergson lays special emphasis on the distinction between the reversible time of physics, in which nothing new happens, and the irreversible time of evolution and biology, in which there is always something new (Bergson, 1910). In spite of this, time in general systems theory is similar to time in classical physics, namely, either all models of systems in general systems theory are still based on the principle of absolute (global) time or time is not explicitly expressed in these models. At the same time, there are many systems, in which it is unfeasible to introduce and preserve global time. For instance, it is proved that clock synchronization becomes impossible under definite conditions (Lamport, 1984; Dolev, et al, 1986; Fischer, et al, 1985; Attiya and Ellen, 2014).This precludes introduction of global time. As a result, in some systems, only local time (local time scale) can be treated in a consistent way. In addition, there are systems, in which global time (global time scale) coexists with a variety of local times (local time scales). Moreover, often these different times and time scales cannot be synchronized. All systems with these properties, which we call concurrent systems, cannot be portrayed by conventional models in general systems theory. The goal of this paper is to construct more advanced than utilized now models of multi-agent distributed systems using the concept of local time, which exists and can be different in distinct components and parts of real systems according to the system theory of time (Burgin, 1992; 2002). These models provide descriptions and tools for exploration not only of classical systems with one global time but also of relativistic and concurrent systems, which can multiplicities of time. It is interesting that absence of global time results in nonexistence of global states in a multicomponent system due to the concurrent functioning of the components and parts. As a result, time becomes multidimensional and demands specific unconventional mathematical structures for its representation. 3

In addition, exploring and modeling systems with a variety of independent and incoherent local times (local time scales), we come to the concepts of an observer, observation, synchronization and coor

This content is AI-processed based on ArXiv data.