TY - GEN
T1 - Judgment aggregation with abstentions under voters' hierarchy
AU - Jiang, Guifei
AU - Zhang, Dongmo
AU - Perrussel, Laurent
PY - 2014
Y1 - 2014
N2 - Similar to Arrow's impossibility theorem for preference aggregation, judgment aggregation has also an intrinsic impossibility for generating consistent group judgment from individual judgments. Removing some of the pre-assumed conditions would mitigate the problem but may still lead to too restrictive solutions. It was proved that if completeness is removed but other plausible conditions are kept, the only possible aggregation functions are oligarchic, which means that the group judgment is purely determined by a certain subset of participating judges. Instead of further challenging the other conditions, this paper investigates how the judgment from each individual judge affects the group judgment in an oligarchic environment. We explore a set of intuitively demanded conditions under abstentions and design a feasible judgment aggregation rule based on the agents' hierarchy. We show this proposed aggregation rule satisfies the desirable conditions. More importantly, this rule is oligarchic with respect to a subset of agenda instead of the whole agenda due to its literal-based characteristics.
AB - Similar to Arrow's impossibility theorem for preference aggregation, judgment aggregation has also an intrinsic impossibility for generating consistent group judgment from individual judgments. Removing some of the pre-assumed conditions would mitigate the problem but may still lead to too restrictive solutions. It was proved that if completeness is removed but other plausible conditions are kept, the only possible aggregation functions are oligarchic, which means that the group judgment is purely determined by a certain subset of participating judges. Instead of further challenging the other conditions, this paper investigates how the judgment from each individual judge affects the group judgment in an oligarchic environment. We explore a set of intuitively demanded conditions under abstentions and design a feasible judgment aggregation rule based on the agents' hierarchy. We show this proposed aggregation rule satisfies the desirable conditions. More importantly, this rule is oligarchic with respect to a subset of agenda instead of the whole agenda due to its literal-based characteristics.
KW - artificial intelligence
KW - intelligent agents (computer software)
KW - judgment aggregation
KW - network security
UR - http://handle.uws.edu.au:8081/1959.7/uws:28680
UR - http://www.prima2014.org/
M3 - Conference Paper
SN - 9783319131900
SP - 341
EP - 356
BT - PRIMA 2014: Principles and Practice of Multi-Agent Systems: 17th International Conference, Gold Coast, QLD, Australia, December 1-5, 2014, Proceedings
PB - Springer
T2 - International Conference on Principles and Practice of Multi-agent Systems
Y2 - 1 December 2014
ER -