[DBWorld] CfP: Fair and Explainable Decision Support Systems

Call for Papers: Feature Issue on Fair and Explainable Decision Support Sys=
tems =


Guest Editors:
-Miguel Couceiro, University of Lorraine, CNRS, Loria (miguel.couceiro@lori=
a.fr)
-Luis Gal=C3=A1rraga, INRIA Rennes ([email protected]) =


Motivation:
Algorithmic decisions are now being employed on a daily basis, and carried =
out by models that are trained on past experiences  and data by Machine Lea=
rning (ML) processes that may be complex and opaque. This lack of transpare=
ncy and the right of explanations of the outcomes of such decision support =
raises several concerns given the critical impact that such decisions may h=
ave on individuals or on society as a whole. Well known examples include de=
cision support systems for loan grants, terrorism detection, prediction of =
criminal recidivism, and many other activities with social and economical i=
mpact on society. In addition to opaqueness issues, many of such decision s=
upport systems have been shown to be biased and leading to outcomes that ca=
n be discriminatory and unfair, which in turn are  notions that remain subj=
ective and dependent both on the empirical scenario, context and decision t=
ask. =


Most of fairness notions focus on the outcomes of the decision process, and=
 they are inspired by several anti-discrimination efforts that aim to ensur=
e that unprivileged groups (e.g. racial minorities) are treated fairly. As =
such, the problem of improving algorithmic fairness can be posed as an opti=
misation one. However, certain fairness dimensions do not fit into this set=
ting, e.g., fairness through unawareness and counterfactuals, and they rais=
e a number of challenges for theorists, researchers and practitioners.

This brings us to the underlying motivation of this Feature Issue that aims=
 at collecting contributions that focus on the various dimensions of algori=
thmic fairness, both from foundational and application perspectives. This r=
anges from papers that suggest frameworks to model fairness, to  address an=
d tackle unfairness, as well as those that propose different aspects in emp=
irical scenarios such as formalization of fairness issues in different appl=
ications (from decision making, operations research, resource allocation an=
d policy making). Contributions dealing with different data-types, e.g., ta=
bular, sequential, textual and other complex data such as graphs, are parti=
cularly welcome.

Contents:
We welcome contributions in the form of state-of-the-art original research =
papers, in the form of position papers that establish bridges between diffe=
rent frameworks, or discussion papers that highlight emerging trends in the=
 topics outlined above. New methodologies, algorithmic tools and implementa=
tions are also within the scope of this Feature Issue.  =


Schedule:
Prospective authors can contact the guest editors with an extended abstract=
 (1.5 pages max, A4 size) of a proposed
paper via e-mail (miguel.couceiro-/[email protected], [email protected]) before=
 submitting the full paper. Submission of full
papers to the Feature Issue is through the electronic submission system: ww=
w.editorialmanager.com/ejdecp,
selecting article type SI:Fair-Decisions.

Important dates:
=E2=80=A2 August 31, 2021: Extended abstract (at least, submission intentio=
n)
=E2=80=A2 December 15, 2021: Submission of full papers
=E2=80=A2 March 31, 2022: Notification (1st round)
=E2=80=A2 June 30, 2022: Revision due
=E2=80=A2 Summer 2022 Publication of Feature issue
_______________________________________________
Please do not post msgs that are not relevant to the database community at =
large.  Go to www.cs.wisc.edu/dbworld for guidelines and posting forms.
To unsubscribe, go to https://lists.cs.wisc.edu/mailman/listinfo/dbworld