A Bayesian approach to norm identification
Cranefield, Stephen; Meneguzzi, Felipe; Oren, Nir; Savarimuthu, Bastin Tony Roy

View/ Open
Cite this item:
Cranefield, S., Meneguzzi, F., Oren, N., & Savarimuthu, B. T. R. (2015). A Bayesian approach to norm identification (Technical Report). Retrieved from http://hdl.handle.net/10523/5476
Permanent link to OUR Archive version:
http://hdl.handle.net/10523/5476
Abstract:
When entering a system, an agent should be aware of the obligations and prohibitions (collectively norms) that will affect it. Existing solutions to this norm identification problem make use of observations of either other's norm compliant, or norm violating, behaviour.
However, they assume an extreme situation where norms are typically violated, or complied with.
In this paper we propose a Bayesian approach to norm identification which operates by learning from both norm compliant and norm violating behaviour. By utilising both types of behaviour, we not only overcome a major limitation of existing approaches, but also obtain improved performance over the state-of-the-art, allowing norms to be learned with a few observations. We evaluate the effectiveness of this approach empirically and discuss theoretical limitations to its accuracy.
Date:
2015
Research Type:
Technical Report
Languages:
English
Collections
- Technical Report [41]
- Information Science [497]