Self-Organising Error Detection and Correction in Open Multi-agent Systems

A key feature of open systems is the expectation of non-compliant behaviour, which may be due to accident, necessity, or malice. Therefore, a system requirement for sustainable operation is to deal with non-compliance through error detection and correction according to type and severity. This paper presents a formal model of a self-organising system of `retributive justice' that encapsulates monitoring, enforcement, and conflict-resolution. Multi-agent simulation is used to evaluate the performance of this system, whereby agents play a variant of the linear public good game and use a behaviour observation framework to adapt their decision-making. Experimental results show that although self-organisation of retributive justice alone cannot guarantee sustainability in an open system and there seems to be no `ideal' system, it can improve the utility of the collective and diminish the likelihood of non-compliance.

[1]  Ilana Ritov,et al.  Winning a competition predicts dishonest behavior , 2016, Proceedings of the National Academy of Sciences.

[2]  Marina De Vos,et al.  I-ABM: combining institutional frameworks and agent-based modelling for the design of enforcement policies , 2013, Artificial Intelligence and Law.

[3]  R. M. Hutchins.  Great Books of the Western World , 1952 .

[4]  Alexander Artikis.  Dynamic specification of open agent systems , 2012, J. Log. Comput..

[5]  Marek J. Sergot,et al.  A Formal Characterisation of Institutionalised Power , 1996, Log. J. IGPL.

[6]  B. Monin,et al.  Victim entitlement to behave selfishly. , 2010, Journal of personality and social psychology.

[7]  Jeremy V. Pitt,et al.  In praise of forgiveness: Ways for repairing trust breakdowns in one-off online interactions , 2008, Int. J. Hum. Comput. Stud..

[8]  Kevin M. Carlsmith,et al.  Psychological Aspects of Retributive Justice , 2007 .

[9]  Simon Gaechter,et al.  Conditional cooperation: Behavioral regularities from the lab and the field and their policy implications , 2006 .

[10]  Dídac Busquets,et al.  The pursuit of computational justice in open systems , 2015, AI & SOCIETY.

[11]  Jeremy V. Pitt,et al.  Provision and Appropriation of Common-Pool Resources without Full Disclosure , 2012, PRIMA.

[12]  Dídac Busquets,et al.  Procedural Justice and 'Fitness for Purpose' of Self-organising Electronic Institutions , 2013, PRIMA.

[13]  G. Brady.  Governing the Commons: The Evolution of Institutions for Collective Action , 1993 .

[14]  Dídac Busquets,et al.  System modeling: Principled operationalization of social systems using Presage2 , 2014, Modeling and Simulation-Based Systems Engineering Handbook.

[15]  Alexander Artikis,et al.  Self-Governance by Transfiguration: From Learning to Prescription Changes , 2014, 2014 IEEE Eighth International Conference on Self-Adaptive and Self-Organizing Systems.