[bull-ia] Post-doc Position (12 months) – Chaire IA BE4musIA (Belief Change for

 

We are offering a post-doc=
position in artificial intelligence. 

This posit=
ion is in relation with the BE4musIA project (a French =C2=ABChaire IA=C2=
=BB project) that is dedicated to the study of tools coming from belief cha=
nge theory, and more generally from knowledge representation and reasoning,=
for improving multi-source information analysis.

&nbs=
p;

** BE4musIA project in short

&=
nbsp;

The aim of the BE4musIA project is to =
use and develop tools coming from knowledge representation and reasoni=
ng (KR) in order to perform an analysis of pieces of information comin=
g from different sources. The idea is to be able to produce at the sam=
e time an assessment of the reliability of the sources and a cohe=
rent view of the world that takes into account the received pieces of infor=
mation.

This problem of trying to find =
the most plausible beliefs from different sources providing conflicting pie=
ces of information is quite a frequent pattern. This is typically the =
case in applications in which one works with several sensors (that are=
typically not totally reliable), when one receives information f=
rom different agents (while some agents can be non-reliable, or, worse&nbsp=
;than that, with some agents being enemies trying to mislead you), or =
when one query several databases in order to answer to a query of a us=
er.
 

Our aim is to study the basic=
abstract problem, using tools mainly coming from the knowledge repres=
entation and reasoning (KR) area, in particular methods coming from belief =
revision and belief merging, that formalize rational belief chang=
e, as well as inconsistency measures, which allows us to measure the e=
xtent to which certain pieces of information are in conflict with each=
other.

 

** Topics for=
the position

 

We welcome a=
pplications in any subject related to the project, in particular: