Synthetic design of a social regressor and its implementation using Knowledge Request-Broker Architecture

By thinking of a society as a cognitive system, this paper proposes a theoretical framework for artificial social intelligence. This framework is implemented using Knowledge Request-Broker Architecture, and its applicability is experienced in the field of system identification by letting a society of self-interested agents, with various skills, to identify a system based on given training data. To develop a theory about how social intelligence works, two questions were asked. First, “What was changed in time that made less intelligent societies to become more intelligent?”, and second “what drives individuals to take part in social productiveness?”. The framework reported in this paper is inspired by the answers given to these question through the study of the emergence of human civilizations, and putting it beside computer and knowledge engineering concepts. The experiments with this system show how social ensemble of simple autonomous agents can solve harder problems, while demonstrating the practicality of the proposed framework and its associated API.

[1]  Linda A. Walton,et al.  In the balance : themes in global history , 1998 .

[2]  Robert A. Jacobs,et al.  Hierarchical Mixtures of Experts and the EM Algorithm , 1993, Neural Computation.

[3]  W. Heath.  The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies , 2008 .

[4]  Thomas Stützle,et al.  Heterogeneous particle swarm optimizers , 2009, 2009 IEEE Congress on Evolutionary Computation.

[5]  Scott E. Page,et al.  Problem Solving by Heterogeneous Agents , 2001, J. Econ. Theory.

[6]  Takao Terano,et al.  A Knowledge Request-Broker Architecture for Development of Artificial Social Intelligence , 2010, 2010 Second International Conference on Computational Intelligence, Modelling and Simulation.

[7]  Timothy W. Finin,et al.  KQML as an agent communication language , 1994, CIKM '94.

[8]  G. Reeke.  Marvin Minsky, The Society of Mind , 1991, Artif. Intell..

[9]  San Cristóbal Mateo,et al.  The Lack of A Priori Distinctions Between Learning Algorithms , 1996 .

[10]  David H. Wolpert,et al.  The Lack of A Priori Distinctions Between Learning Algorithms , 1996, Neural Computation.

[11]  Michael Winikoff,et al.  JACKTM Intelligent Agents: An Industrial Strength Platform , 2005, Multi-Agent Programming.

[12]  Reid G. Smith,et al.  The Contract Net Protocol: High-Level Communication and Control in a Distributed Problem Solver , 1980, IEEE Transactions on Computers.

[13]  Ciarán O'Leary,et al.  Building a Hybrid Society of Mind Using Components from Ten Different Authors , 2003, ECAL.

[14]  Jonathan Ozik,et al.  Visual agent-based model development with repast simphony. , 2007 .

[15]  Finn E. Kydland.  Heterogeneous agents in quantitative aggregate economic theory , 1994 .

[16]  Milind Tambe,et al.  Towards Heterogeneous Agent Teams , 2001, EASSS.