[bull-ia] Séminaire GDR IA de Meltem Ozturk: 15 décem

Content-Type: text/html; charset= »utf-8″

C=E2=80=99est demain:


Chers coll=C3=A8=

Le prochain s=C3=A9minaire mensuel du GD=
R IA, dont vous trouverez
les d=C3=A9tails ci-dessous, sera d=
onn=C3=A9 par Meltem Ozturk
(https://www.lamsade.dauphine.fr/~ozturk/). Pour rappel, le but de ces
expos=C3=A9s est de
mettre en lumi=C3=A8re, au sein d’un expos=C3=A9
ible =C3=A0 la plus grande partie de la communaut=C3=A9, les domaines
vari=C3=A9s couverts par le GDR et ses activit=C3=A9s.

Lien vers la salle virtuelle: https://utc-fr.zoom.us/j/89907120037

Les informations sur les s=C3=A9minaires s=
e trouvent sur la page du GDR IA:
Date: le 15/12 =C3=A0 11h

Some axiomatic and algorithmic perspectives on the social
nking problem and its relation with power indices

R=C3=A9sum=C3=A9: "In this talk, I will present some recent stud=
ies on social
ranking from axiomatic and computational point =
of views and show it’s
relation with power indices.

In many domains, a number of works have been devoted t=
o ranking
individuals/objects based on the performance of the=
groups formed by
them. Besides game theory, arguably the fie=
ld which has dealt with
this question the most, we can think =
of networks (the influence of
someone in a social network), b=
elief merging/revision (the
responsibility of a formula in th=
e inconsistency of a belief base),
multicriteria decision aid=
ing (the impact and synergy of some
criteria), machine learni=
ng (selecting best features), argumentation
(influence of an =
argument in a debate), etc.

Power indices (suc=
h as Shapley or Banzhaf indices), introduced in
cooperative g=
ame theory, deal with this problem in a cardinal way. But
y real-life situations do not fit this framework, as in particular
we may only have ordinal information about groups of agents. Socialranking proposes a more flexible theory of cooperative interac=
situations considering only ordinal comparisons between =

In this talk, I will present different=
social ranking rules, show
their differences based on their =
axiomatizations. I will also point
out some computational res=
ults (such as manipulability of such rules).
This ordinal app=
roach being quite recent, I will conclude by some open

Nous tenons =C3=A0 remercier tous l=
es GT qui nous ont fait remonter des
propositions de s=C3=A9m=
inaires, et tous les GT qui le feront.

Au plai=
sir de vous "voir" bient=C3=B4t.

olas et S=C3=A9bastien (D.)