| 研究実績の概要 |
Our work lies in the field of Opinion and Belief Propagation in Social Networks, and we mainly use the framework of Belief Revision Games we proposed in a AAAI'15 paper. In FY2018, we focused on two topics:
- We introduced a new class of belief change operators, named promotion operators. The aim of these operators is to enhance the acceptation of a formula representing a new piece of information. We gave postulates for these operators and provided a representation theorem in terms of minimal change. We also showed that this class of operators is a very general one, since it captures as particular cases belief revision, commutative revision, and (essentially) belief contraction. This operation of promotion is used to update the beliefs of an agent in social network. This is useful when one needs to change the opinion of an agent in the network so as to avoid the propagation of fallacious information over all the network. We have published a paper on this topic to the KR'18 conference, the top world-leading conference in the field of Knowledge Representation and Reasoning.
- We have also worked on the notion of public announcement learning. More precisely, we considered the problem of identifying the change formula in a belief revision scenario: given that an unknown public announcement led a set of agents to revise their beliefs and given the prior beliefs and the revised beliefs of the agents, what can be said about the announcement? We got a full paper accepted to IJCAI'19, the world leading conference in Artificial Intelligence.
|
| 今後の研究の推進方策 |
The plan for FY2019 will be to present our paper accepted to IJCAI'19, and to work on the notion of evaluation of the reliability of information sources. This will be done by taking advantage of both machine learning methods (e.g., ensemble methods) and belief merging theory. The goal is to learn iteratively a weight for each datasource, representing how reliable the source is. A paper will be submitted to AAAI'20.
|
