Intelligent Systems 

Discussion group on Entropy and Self-organisation in Multi-Agent Systems

Self-Assembly,  Amorphous Computing,  Artificial life

Radhika Nagpal. Programmable self-assembly using biologically-inspired multiagent control. Proceedings of the First International Joint Conference on Autonomous Agents and Multi-Agent Systems, (AAMAS'02), 2002.

       Abstract. This paper presents a programming language that species a robust process for shape formation on a sheet of identically programed agents, by combining local organization primitives from epithelial cell orphogenesis and Drosophila cell differentiation with combination rules from geometry. This work represents a significantly different approach to the design of self-organizing systems: the desired global shape is specified using an abstract geometry-based language, and the agent program is directly compiled from the global specification. The resulting self-assembly process is extremely reliable in the face of random agent distributions, random agent death and varying agent numbers, without relying on global coordinates or centralized control. 

Steen Rasmussen, Nils A. Baas, Bernd Mayer, Martin Nilsson and Michael W. Olesen.  Ansatz for Dynamical Hierarchies. Artificial Life, volume 7, number 4, 2001.

       Abstract. Complex, robust functionalities can be generated naturally in at least two ways: by the assembly of structures and by the evolution of structures. This work is concerned with spontaneous formation of structures. We define the notion of dynamical hierarchies in natural systems and show the importance of this particular kind of organization for living systems. ... Our framework is then applied to a concrete and very simple formal, physicochemical, dynamical hierarchy involving water and monomers at level one, polymers and water at level two, and micelles (polymer aggregates) and water at level three. Formulating this system as a simple two-dimensional molecular dynamics (MD) lattice gas allows us within one dynamical system to demonstrate the successive emergence of two higher levels (three levels all together) of robust structures with associated properties. Second, we demonstrate how the framework for dynamical hierarchies can be used for realistic (predictive) physicochemical simulation of molecular self-assembly and self-organization processes. We discuss the detailed process of micellation using the three-dimensional MD lattice gas. Finally, from these examples we can infer principles about formal dynamical hierarchies. We present an ansatz for how to generate robust, higher-order emergent properties in formal dynamical systems that is based on a conjecture of a necessary minimal complexity within the fundamental interacting structures once a particular simulation framework is chosen.

Barry McMullin, Francisco J. Varela. Rediscovering Computational Autopoiesis (ECAL-97).

       Abstract. This paper summarises some initial empirical results from a new computer model (artificial chemistry) which exhibits spontaneous emergence and persistence of autopoietic organisation. The model is based on a system originally presented by Varela, Maturana and Uribe [11]. In carrying out this re-implementation it was found that an additional interaction (chain-based bond inhibition), not documented in the original description by Varela et al., is critical to the realisation of the autopoietic phenomena. This required interaction was re-discovered only following careful examination of (unpublished) source code for an early version of the original model. The purpose of the paper is thus twofold: firstly to identify and discuss this previously undocumented, but essential, interaction; and secondly to argue, on the basis of this particular case, for the importance of exploiting the emerging technologies which support publication of completely detailed software models (in addition, of course, to conventional publication of summary experimental results).

Lisa J. K. Durbeck and Nicholas J. Macias. Defect-tolerant, fine-grained parallel testing of a Cell Matrix (SPIE ITCom 2002).

       Abstract. A fault testing methodology for a cell-based self configurable hardware platform (the Cell Matrix) is described. Background on the Cell Matrix is given, including its amenability to use despite the presence of manufacturing defects. The ability of cells within the Cell Matrix to isolate faulty regions is also described. A method for testing individual cells, based on an external test driver, is discussed. The benefits of locating this test driver inside the device under test are explained. A method is described for efficient, autonomous, robust creation of a network of self-testing structures (called Supercells) for parallel implementation and execution of this test driver. Sample tests are described, and their results are given, demonstrating the effectiveness and robustness of the testing methodology. A discussion of the research, including conclusions, is presented. Plans for future work are discussed.

Frank Schweitzer, Benno Tilch. Self-Assembling of Networks in an Agent-Based Model. Physical Review E, 66, 026113 (1-9), 2002.

       Abstract. We propose a model to show the self-assembling of network-like structures between a set of nodes without using preexisting positional information or long-range attraction of the nodes. The model is based on Brownian agents that are capable of producing different local (chemical) information and respond to it in a non-linear manner. They solve two tasks in parallel: (i) the detection of the appropriate nodes, and (ii) the establishment of stable links between them. We present results of computer simulations that demonstrate the emergence of robust network structures and investigate the connectivity of the network by means of both analytical estimations and computer simulations.

Contact:

Dr Mikhail Prokopenko
Tel : 61 (02) 9325 3264
Fax: 61 (02) 9325 3200
mikhail.prokopenko@csiro.au