TY - GEN
T1 - Imperfect norm enforcement in stochastic environments:
T2 - an analysis of efficiency and cost tradeoffs
AU - Fagundes, Moser Silva
AU - Meneguzzi, Felipe
AU - Ossowski, Sacha
PY - 2014/11/12
Y1 - 2014/11/12
N2 - In heterogeneous multiagent systems, agents might interfere with each other either intentionally or unintentionally, as a side-effect of their activities. One approach to coordinating these agents is to restrict their activities by means of social norms whose compliance ensures certain system properties, or otherwise results in sanctions to violating agents. While most research on normative systems assumes a deterministic environment and norm enforcement mechanism, we formalize a normative system within an environment whereby agent actions have stochastic outcomes and norm enforcement follows a stochastic model in which stricter enforcement entails higher cost. Within this type of system, we analyze the tradeoff between norm enforcement efficiency (measured in number of norm violations) and its cost considering a population of norm-aware self-interested agents capable of building plans to maximize their expected utilities. Finally, we validate our analysis empirically through simulations in a representative scenario.
AB - In heterogeneous multiagent systems, agents might interfere with each other either intentionally or unintentionally, as a side-effect of their activities. One approach to coordinating these agents is to restrict their activities by means of social norms whose compliance ensures certain system properties, or otherwise results in sanctions to violating agents. While most research on normative systems assumes a deterministic environment and norm enforcement mechanism, we formalize a normative system within an environment whereby agent actions have stochastic outcomes and norm enforcement follows a stochastic model in which stricter enforcement entails higher cost. Within this type of system, we analyze the tradeoff between norm enforcement efficiency (measured in number of norm violations) and its cost considering a population of norm-aware self-interested agents capable of building plans to maximize their expected utilities. Finally, we validate our analysis empirically through simulations in a representative scenario.
U2 - 10.1007/978-3-319-12027-0_42
DO - 10.1007/978-3-319-12027-0_42
M3 - Published conference contribution
VL - 8864
T3 - Lecture Notes in Computer Science
SP - 523
EP - 535
BT - Advances in Artificial Intelligence - IBERAMIA 2014
PB - Springer
ER -