Youssef Hamadi, Microsoft Research Cambridge,
Eric Monfroy, Université de Nantes/UTFSM Valparaiso,
Frederic Saubion, Université d’Angers.
Recent progresses in the processing of combinatorial problems have demonstrated that search algorithms can become extremely efficient when they take advantage of previous attempts to drive their exploration. For instance in modern DPLLs, the collect of conflicts feeds the variable selection heuristic, and the quality of unit propagation controls the use of the restart strategy. These implicit uses of closed control loops have been empirically discovered by the SAT community. We believe that a more principled and autonomous approach for search efficiency has to be started in CSP and Constraint Programming. The purpose of this special issue is to present recent attempts in the design of autonomous solvers, therefore works coming from different communities are especially welcomed.
Typical topics include, but are not limited to:
Self configuration of search
Self optimization of search
Learning for search
Papers will be reviewed by an international committee according to the following criteria:
Pertinence: does the paper describe methods or theories concerning the topics of the special issue?
Scientific quality: does the paper clearly identify a scientific problem, document the state of the art and demonstrate an original technique or method that resolves the problem?
Impact: is the method/model likely to be adopted for the design or evaluation of solvers?
Generality: can the method/model be used for a variety of solvers?
Innovation: does the method/model demonstrate an improvement in the current state of the art?
All the papers should be full journal length versions and follow the guidelines set out by CPL. All the papers will be peer reviewed following the CPL reviewing procedures. The guide for authors can be found here: http://www.cs.brown.edu/people/pvh/CPL/format/authors-guide.html
Authors should send a pdf file to Frederic.Saubion@univ-angers.fr
Submission deadline: 15th of June 2008
Publication of the special issue: December 2008
Constraint Programming Letters (CPL) provides an international forum for the electronic publication of high-quality scholarly articles on constraint programming. All published papers are freely available online. The goal of CPL is to promote and nurture constraint programming research, report on its successful applications, and encourage cross-fertilization with neighbouring areas.