BLANC - Blanc

Génération Aléatoire : Modèles, Méthodes et Algorithmes – GAMMA

Submission summary

The goal of this project, "Random Generation: Models, Methods and Algorithms", is to develop a set of fundamental methods and algorithms for the random generation of complex combinatorial structures. Such objects appear in a variety of applications, especially in computer science. Random generation plays an important role in domains dealing with enormous volumes of data, for example in interaction networks (computer, sociological, biological networks); in this case, random generation allows the comparison of models and the evaluation of their adequacy for real data. Random testing and model checking are two additional fields of application for random generation. One of the main issues there arises when choosing appropriate data for testing, and designing efficient methods for fast generation of this data. Our goal is to study all the steps leading to the realization of efficient generators. More precisely, the aim is first to study and apprehend complex combinatorial objects such as graphs, maps, automata or generalized partitions. Then, this analysis enables us to give recognition algorithms for our objects, useful for generation with rejection for instance, and perhaps efficient transformation algorithms between our complex objects and simpler ones, in order to exhaustively or randomly generate them. Another goal of our project is the development of the algorithmics of random generation and, in particular, algorithms related to Boltzmann model. We want to extend generic algorithms to - make multicriteria random generation, - take into account new specifications, - control numerical approximations and the loss of uniformity induced, - use discrete rather than floating point versions of these algorithms. Our project is mainly based on the productive synergy between combinatorial and algorithmic tools requiered to attain our objectives.

Project coordination

Frédérique BASSINO (UNIVERSITE DE PARIS XIII)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

UNIVERSITE DE PARIS XIII

Help of the ANR 232,000 euros
Beginning and duration of the scientific project: - 36 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter