Societal Implicit Memory and his Speed on Tracking
Dynamic Extrema
64. Vitorino Ramos,
Carlos Fernandes, Agostinho C. Rosa, Societal Implicit Memory and his
Speed on Tracking Extrema over Dynamic Environments using
Self-Regulatory Swarms, to appear soon.
Fig. - (Left)
A 3D toroidal fast changing landscape describing a Dynamic Optimization
(DO) Control Problem (8 frames in total). (Rigth) A self-organized swarm emerging a
characteristic flocking migration behaviour surpassing in intermediate
steps some local optima over the 3D toroidal landscape (left),
describing a Dynamic Optimization (DO) Control Problem. Over each
foraging step, the swarm self-regulates his population and keeps
tracking the extrema (44 frames in total).
PDF file: paper (6540 Kb)
Abstract: In order to overcome difficult dynamic optimization and environment extrema tracking problems, we propose a Self-Regulated Swarm (SRS) algorithm which hybridizes the advantageous characteristics of Swarm Intelligence as the emergence of a societal environmental memory or cognitive map via collective pheromone laying in the landscape (properly balancing the exploration/exploitation nature of the search strategy), with a simple Evolutionary mechanism that trough a direct reproduction procedure linked to local environmental features is able to self-regulate the above exploratory swarm population, speeding it up globally. In order to test his adaptive response and robustness, we have recurred to different dynamic multimodal complex functions as well as to Dynamic Optimization Control (DOC) problems. Measures were made for different dynamic settings and parameters such as, environmental upgrade frequencies, landscape changing speed severity, type of dynamic (linear or circular), and to dramatic changes on the algorithmic search purpose over each test environment (e.g. shifting the extrema). Finally, comparisons were made with traditional Genetic Algorithms (GA) as well as with more recently proposed Co-Evolutionary approaches. SRS, were able to demonstrate quick adaptive responses, while outperforming the results obtained by the other approaches. Additionally, some successful behaviors were found: SRS was able not only to achieve quick adaptive responses, as to maintaining a number of different solutions, while adapting to new unforeseen extrema; the possibility to spontaneously create and maintain different subpopulations on different peaks, emerging different exploratory corridors with intelligent path planning capabilities; the ability to request for new agents over dramatic changing periods, and economizing those foraging resources over periods of stabilization. Finally, results prove that the present SRS collective swarm of bio-inspired agents is able to track about 65% of moving peaks traveling up to ten times faster than the velocity of a single ant composing that precise swarm tracking system. This emerged behavior is probably one of the most interesting ones achieved by the present work.
º Wawrzyniak, D., Obuchowicz, A., "Evolutionary Algorithm with Forced Variation in Multi-Dimensional Non-Stationary Environment", in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4967, LNCS, pp. 600-607, Springer, 2008.
º Siphesihle M. Zuma, Matthew O. Adigun, "CACIP: a Pattern for Interfacing Components in a Context-aware Mobile Environment", in Proc. of the 17th IASTED Int. Conf. on Modelling and Simulation, pp. 416-423, Acta Press, Montreal, Canada, 2006.
º Laszlo P.K. Mahanti and S. Banerjee, "Automated Testing in Software Engineering: Using Ant Colony and Self-Regulated Swarms", in R. Wamkeue (Ed.), MS 2006, Modelling and Simulation Int. Conf., ACTA Press, ISBN: 0-88986-592-2, Montreal, QC, Canada, May 2006.
Related Works:
61. On Self-Regulated Swarms, Societal Memory, Speed and Dynamics.
58. On Ants, Bacteria and Dynamic Environments.
PDF file: paper (6540 Kb)
Abstract: In order to overcome difficult dynamic optimization and environment extrema tracking problems, we propose a Self-Regulated Swarm (SRS) algorithm which hybridizes the advantageous characteristics of Swarm Intelligence as the emergence of a societal environmental memory or cognitive map via collective pheromone laying in the landscape (properly balancing the exploration/exploitation nature of the search strategy), with a simple Evolutionary mechanism that trough a direct reproduction procedure linked to local environmental features is able to self-regulate the above exploratory swarm population, speeding it up globally. In order to test his adaptive response and robustness, we have recurred to different dynamic multimodal complex functions as well as to Dynamic Optimization Control (DOC) problems. Measures were made for different dynamic settings and parameters such as, environmental upgrade frequencies, landscape changing speed severity, type of dynamic (linear or circular), and to dramatic changes on the algorithmic search purpose over each test environment (e.g. shifting the extrema). Finally, comparisons were made with traditional Genetic Algorithms (GA) as well as with more recently proposed Co-Evolutionary approaches. SRS, were able to demonstrate quick adaptive responses, while outperforming the results obtained by the other approaches. Additionally, some successful behaviors were found: SRS was able not only to achieve quick adaptive responses, as to maintaining a number of different solutions, while adapting to new unforeseen extrema; the possibility to spontaneously create and maintain different subpopulations on different peaks, emerging different exploratory corridors with intelligent path planning capabilities; the ability to request for new agents over dramatic changing periods, and economizing those foraging resources over periods of stabilization. Finally, results prove that the present SRS collective swarm of bio-inspired agents is able to track about 65% of moving peaks traveling up to ten times faster than the velocity of a single ant composing that precise swarm tracking system. This emerged behavior is probably one of the most interesting ones achieved by the present work.
Keywords: Dynamic Optimization, Dynamic Optimal Control problems, Swarm Intelligence, Self-Organization, Societal Implicit Memory, Stigmergy.
Cited by:º Wawrzyniak, D., Obuchowicz, A., "Evolutionary Algorithm with Forced Variation in Multi-Dimensional Non-Stationary Environment", in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4967, LNCS, pp. 600-607, Springer, 2008.
º Siphesihle M. Zuma, Matthew O. Adigun, "CACIP: a Pattern for Interfacing Components in a Context-aware Mobile Environment", in Proc. of the 17th IASTED Int. Conf. on Modelling and Simulation, pp. 416-423, Acta Press, Montreal, Canada, 2006.
º Laszlo P.K. Mahanti and S. Banerjee, "Automated Testing in Software Engineering: Using Ant Colony and Self-Regulated Swarms", in R. Wamkeue (Ed.), MS 2006, Modelling and Simulation Int. Conf., ACTA Press, ISBN: 0-88986-592-2, Montreal, QC, Canada, May 2006.
Related Works:
70. Computational
Chemotaxis
in Ants and Bacteria
over Dynamic
Environments.
69. Binary Ant Algorithm.
63. Social
Cognitive Maps, Swarm
Collective Perception and Distributed Search on Dynamic Landscapes.
61. On Self-Regulated Swarms, Societal Memory, Speed and Dynamics.
58. On Ants, Bacteria and Dynamic Environments.