Time filter

Source Type

Cazenave T.,CNRS Laboratory for the Analysis and Modeling of Decision Systems
Theoretical Computer Science | Year: 2016

Monte Carlo Tree Search (MCTS) is the state of the art algorithm for General Game Playing (GGP). We propose to learn a playout policy online so as to improve MCTS for GGP. We also propose to learn a policy not only using the moves but also according to the features of the moves. We test the resulting algorithms named Playout Policy Adaptation (PPA) and Playout Policy Adaptation with move Features (PPAF) on Atarigo, Breakthrough, Misere Breakthrough, Domineering, Misere Domineering, Knightthrough, Misere Knightthrough and Nogo. The experiments compare PPA and PPAF to Upper Confidence for Trees (UCT) and to the closely related Move-Average Sampling Technique (MAST) algorithm. © 2016. Source

Mayag B.,CNRS Laboratory for the Analysis and Modeling of Decision Systems
Communications in Computer and Information Science | Year: 2016

In a context of Multiple Criteria Decision Aid, we present a decision model explaining some French hospitals rankings in weight loss surgery. To take into account interactions between medical indicators, we elaborated a model based on the 2-additive Choquet integral. The reference subset, defined during the elicitation process of this model, is composed by some specific alternatives called binary alternatives. To validate our approach, we showed that the proposed 2-additive Choquet integral model is able to approximate the hospitals ranking, in weight loss surgery, published by the French magazine “Le Point” in August 2013. © Springer International Publishing Switzerland 2016. Source

Bouyssou D.,CNRS Laboratory for the Analysis and Modeling of Decision Systems | Marchant T.,Ghent University
Journal of Informetrics | Year: 2016

This paper analyzes from an axiomatic point of view a recent proposal for counting citations: the value of a citation given by a paper is inversely proportional to the total number of papers it cites. This way of fractionally counting citations was suggested as a possible way to normalize citation counts between fields of research having different citation cultures. It belongs to the "citing-side" approach to normalization. We focus on the properties characterizing this way of counting citations when it comes to ranking authors. Our analysis is conducted within a formal framework that is more complex but also more realistic than the one usually adopted in most axiomatic analyses of this kind. © 2015 Elsevier Ltd. Source

Sali M.,University of Paris Dauphine | Giard V.,CNRS Laboratory for the Analysis and Modeling of Decision Systems
Production Planning and Control | Year: 2015

This paper examines the centralised monitoring of an upstream supply chain (USC) in the context of a mass customisation production system. We propose an adaptation of the material requirement planning framework to manage demand uncertainty at each stage of the USC. New analytical relations that exploit structural information on demand beyond the frozen horizon are developed for efficient decision-making and appropriate information sharing among the productive supply chain units. A numerical example to illustrate the implementation of the new analytical relations is provided, and a synthesis of benchmarks performed in the automotive industry is presented. © 2015 Taylor & Francis. Source

Gourves L.,CNRS Laboratory for the Analysis and Modeling of Decision Systems
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

This paper deals with states that are immune to group deviations. Group deviations help the players of a strategic game to escape from undesirable states but they compromise the stability of a system. We propose and analyse a solution concept, called profitable deviation strong equilibrium, which is between two well-known equilibria: the strong equilibrium and the super strong equilibrium. The former precludes joint deviations by groups of players who all benefit. The latter is more demanding in the sense that at least one member of a deviating coalition must be better off while the other members cannot be worst off. We study the existence, computation and convergence to a profitable deviation strong equilibrium in three important games in algorithmic game theory: job scheduling, max cut and singleton congestion game © Springer International Publishing Switzerland 2015. Source

Discover hidden collaborations