Michel Sintzoff

Centres d'intérêt scientifiques / Scientific Interests

• Méthodes de conception de logiciel/ Software design methods
• Conception d'algorithmes/ Algorithm design

Activités extérieures / External activities

IFIP Working Group 2.1 on Algorithmic Languages and Calculi: http://web.comlab.ox.ac.uk/oucl/work/jeremy.gibbons/wg21/

IFIP Working Group 2.3 on Programming Methodology: http://research.microsoft.com/~leino/IFIP-WG2.3/

Articles récents / Recent papers:

• M.Sintzoff,
  Iterative synthesis of winning strategies ensuring invariance and inevitability in discrete-decision games,
  Research Rep., RR 2003-03, Dept of Computing Science and Engineering, Université catholique de Louvain, March 2003.
  
  Abstract - Reactive and hybrid systems can be modeled by games where players make strategic decisions in a temporally discrete manner. Players may use dense- or discrete-time dynamics. It is proposed to restrict the proponent moves by "winning guards" in order to guarantee invariance and inevitability properties. The winning strategy determined by these winning guards should not exclude any initial state from which a winning strategy exists. Sets of such initial states constitute winning regions and are defined by fixed points. Iterates which yield winning guards are obtained by decomposing the iterates which yield winning regions.
  
  The report can obtained from secret@info.ucl.ac.be
  
  
• M.Sintzoff,
  On the design of correct and optimal dynamical systems and games, Information Processing Letters 88 (2003) 59-65.
  
  Abstract - There exist various methods for designing dynamical systems and dynamical games in order to ensure correctness and optimality. In the paper, they are systematically organized as follows. Two variational principles are recalled. Firstly, solutions must be stationary: this leads to necessary conditions and to gradient algorithms. Secondly, solutions, if any, must be optimal or correct; this leads to sufficient conditions and to dynamic-programming algorithms. Methods based on these principles allow to design dynamical systems and games such as control systems, hybrid systems and reactive ones. Time may be discrete or continuous; correctness can be viewed as an abstraction of optimality. The structured presentation of design methods is intended to foster their understanding, integration, cross-fertilization and improvement.
  
  The paper can be obtained from http://www.sciencedirect.com/journal/00200190