Researchers in multi-agent systems (MAS) are interested in how autonomous "software agents" can be designed to interact in an open society of agents (both software and human). Like human society, systems of autonomous software are more effective if agents interact in an orderly and predictable fashion. To achieve this, MAS researchers have borrowed many social and organisational concepts from human society and developed computational counterparts.
In this talk I will discuss my recent and emerging research in this area. In particular, I will discuss how agents can understand human values, the meaning of social commitments, and what it means for autonomous software to be accountable.
Last modified: Wednesday, 01-Aug-2018 14:39:28 NZST
This page is maintained by the seminar list administrator.