MaNEA: A distributed architecture for enforcing norms in open MAS
Introduction
The main applications of Multi-Agent Systems (MAS) support large scale open distributed systems. These systems are characterized by the heterogeneity of their participants; their limited trust; a high uncertainty; and the existence of individual goals that might be in conflict (Artikis and Pitt, 2001). In these scenarios, norms are conceived as an effective mechanism for achieving coordination and ensuring social order; i.e., norms represent an effective tool for regulating the actions of software agents and the interactions among them (López et al., 2006). Most of the proposals on methodologies and guidelines aimed at developing open MAS (Argente et al., 2011b, Dignum et al., 2005) are based on organizational concepts, such as norms. These concepts facilitate the analysis and design of coordination and collaboration mechanisms for MAS. Therefore, norms should be considered in the design and specification of the MAS (Criado et al., 2011b). As pointed out in Castelfranchi (2003), the use of norms in MAS allows better results to be achieved in dynamic and complex environments. Specifically, the fact that agents can violate norms autonomously allows a better adaptation to the environmental changes. Finally, the occurrence of norm violations can evidence the need to adapt the MAS (Bernon et al., 2003). Agent platforms are the software that supports the development and execution of MAS. Thus, norms must be also considered in the design and implementation of agent platforms (Criado et al., 2011b). As a consequence, agent platforms must implement norms in an optimized way, given that in open MAS the internal states of agents are not accessible (Criado et al., 2011a). Therefore, norms cannot be imposed as agent's beliefs or goals, but they must be implemented in the platforms by means of control mechanisms (Grossi et al., 2007).
This paper shows an overview of the most relevant works on norm implementation. This paper considers the main challenges of open MAS and points out the main deficiencies and drawbacks of agent platforms and infrastructures when supporting norms. With the aim of overcoming some of these problems, in this paper a Norm-Enforcing Architecture, known as MaNEA, is proposed. Specifically, MaNEA has been integrated into the Magentix2 platform.1 The Magentix2 platform allows the management of open MAS in a secure and optimized way. Its main objective is to bring agent technology to real domains: business, industry, e-commerce, among others. This goal entails the development of more robust and efficient mechanisms for enforcing norms that control these complex applications.
This paper is organized as follows: Section 2 contains the analysis of the main proposals on norm enforcement; Section 3 describes briefly the Magentix2 platform; Section 4 describes the main components of MaNEA; Section 5 illustrates the performance of MaNEA through a case study; Section 6 contains an evaluation of this architecture; and, finally, Section 7 contains conclusions and future works.
Section snippets
Related work
Most of the proposals on norms for controlling MAS tackle this issue from a theoretical perspective (Boella, 2004, Sergot, 1998). However, there are also works on norms from a computational point of view. These works propose control mechanisms for norms to have an effective influence on agent behaviours (Grossi et al., 2007). These control mechanisms are classified into two categories (Grossi et al., 2007): regimentation mechanisms, which consist of making the violation of norms impossible; and
The Magentix2 platform
Magentix2 is an agent platform for open MAS in which heterogeneous agents interact and organize themselves into Virtual Organizations (VOs) (Foster et al., 2001). VOs are open systems formed by the grouping and collaboration among heterogeneous entities. In VOs there is a separation between form and function that requires defining how behaviour will take place (Ferber et al., 2005). VOs are social entities formed by agents that try to achieve the organizational goals. These agents are organized
Norm-enforcing architecture: MaNEA
The main purpose of MaNEA (Magentix2 Norm-Enforcing Architecture) is to endow the Magentix2 platform with an infrastructure capable of controlling norms in open MAS where unforeseen scenarios may occur. Therefore, the number of agents and the situations that must be controlled through norms may change at runtime. For this reason, MaNEA has been designed as a distributed architecture. Specifically, MaNEA has been distributed over two layers (see Fig. 6):
- •
The highest layer of MaNEA is formed by
Case study
The first step that any engineer must perform to implement a MAS in Magentix2 is to model the MAS based on the VO approach. To facilitate this task, engineers are encouraged to use the GORMAS guidelines (Argente et al., 2011b). Specifically, GORMAS (Guidelines for ORganizational Multi-Agent Systems) proposes a set of activities for the analysis and the design of VOs. These guidelines cover the requirement analysis, the structure design and the design of the organization-dynamics. Software
Evaluation
Bringing agent technology to real domains such as industry or business entails the development of efficient platforms and infrastructures capable of engineering open distributed systems. Efficiency is one of the main requirements that have motivated the design of MaNEA. To this goal, MaNEA is based on a tracing service, which has been implemented following a publish/subscribe metaphor. Traditionally, Norm-Enforcing Architectures have been built using overhearing approaches. Overhearing is
Conclusions and future works
In this paper, we have described a Norm-Enforcing Architecture (MaNEA) that has been developed considering the facilities provided by the Magentix2 platform. The main aim of MaNEA is to overcome problems of existing proposals on norm enforcement. MaNEA enforces norms providing support to those agents that are not endowed with normative reasoning capabilities. In addition, the generation of events for informing about sanctions and rewards allows norm-aware agents to use this information for
Acknowledgement
We acknowledge ITMAS 2011 as the forum in which the main ideas behind this paper were preliminary discussed. This paper was partially funded by the Spanish government under projects CONSOLIDER-INGENIO 2010 CSD2007-00022, TIN2009-13839-C03-01, TIN2012-36586-C03-01 and by the FPU grant AP-2007-01256 awarded to N.Criado. This research has also been partially funded by the Generalitat de Catalunya under the grant 2009-SGR-1434.
References (52)
- et al.
TRAMMASa tracing model for multiagent systems
Eng. Appl. Artif. Intell.
(2011) - et al.
Multi-agent systems in production planning and controlan application to the scheduling of mixed-model assembly lines
Int. J. Prod. Econ.
(2000) Formalising the informal? Dynamic social order, bottom-up social control, and spontaneous normative relations
J. Appl. Logic
(2003)- et al.
Partial identities as a foundation for trust and reputation
Eng. Appl. Artif. Intell.
(2011) - Argente, E., 2008. Gormas: Guidelines for Organization-Based Multiagent Systems. Ph.D. Thesis, Universidad Politécnica...
- et al.
An abstract architecture for virtual organizationsthe THOMAS approach
Knowl. Inf. Syst.
(2011) - Argente, E., Botti, V., Julian, V., 2011b. Gormas: an organizational-oriented methodological guideline for open MAS....
- Artikis, A., Pitt, J., 2001. A formal model of open agent societies. In: Proceedings of the International Conference on...
- Bernon, C., Gleizes, M., Peyruqueou, S., Picard, G., 2003. ADELFE: a methodology for adaptive multi-agent systems...
- Boella, G., van der Torre, L., 2003. Norm governed multiagent systems: the delegation of control to autonomous agents....
Institutional reality and normsspecifying and monitoring agent organizations
Int. J. Coop. Inf. Syst.
Open issues for normative multi-agent systems
AI Commun.
Organization Theory and Design
Rationality and the emotions
Econ. J.
Cited by (20)
The Information Flow Problem in multi-agent systems
2018, Engineering Applications of Artificial IntelligenceCitation Excerpt :An alternative technique for indirect communication is the use of tracing facilities, as it has been recently proposed by Búrdalo et al. (2011). Tracing in multi-agent systems has been traditionally focused on providing human users with debugging or monitoring information (Bellifemine et al., 2007; Collis et al., 1998; Serrano et al., 2012); but some other examples show the usefulness of tracing as a general indirect information scheme available to agents, as in Criado et al. (2013), where norm control in open MAS is performed by using an event-tracing approach. Other approaches in the literature have focused on the analysis of the communication process within multi-agent systems, with the purpose of helping MAS designers by offering tools capable to optimize the communication processes among the designed agents.
BDI reasoning with normative considerations
2015, Engineering Applications of Artificial IntelligenceCitation Excerpt :That is, we can represent partial compliance with a norm as the fraction of utility that an agent can garner by executing a plan in compliance with any CTDs, as opposed to the utility gained by complying with the initial norm. Criado (2012) and Criado et al. (2010, 2013) propose MaNEA (Magentix2 Norm-Enforcing Architecture), a distributed architecture based on the Magentix2 platform (Such et al., 2012), enabling agents to handle uncertainty (of facts about the world), norm salience (i.e., its importance for a given context) and norm relevance (i.e., whether a norm is applicable or not to a given context) when deciding to comply or not with norms. The proposal is very expressive, but designers would face a challenge to properly model probabilities of beliefs, cognitive states, salience and relevance, as these are numeric values whose aggregated/combined effect may be unpredictable.
Classifying sanctions and designing a conceptual sanctioning process model for socio-technical systems
2016, Knowledge Engineering ReviewCommitment-based negotiation semantics for accountability in multi-agent systems
2023, Annals of Mathematics and Artificial IntelligenceNC4OMAS: A Norms-based Approach for Open Multi-Agent Systems Controllability
2022, International Conference on Agents and Artificial IntelligenceNormative Emotional Agents: A Viewpoint Paper
2022, IEEE Transactions on Affective Computing