Olivier Coulon1;2, Isabelle Bloch1, Vincent Frouin2, Jean-Fran cois Mangin2 1 Ecole Nationale Sup erieure des T el ecommunications, Paris, France. 2 Service Hospitalier Fr ed eric Joliot, CEA, Orsay, France. Introduction The problem of brain activation detection in PET activation maps has essentially been approached in terms of statistical analysis (1). Although these analyses have been validat...

We discuss the temporal di erence learning algorithm as applied to approximating the cost to go function of an in nite horizon discounted Markov chain The algorithm we analyze updates parameters of a linear function approximator on line during a single endless traject ory of an irreducible aperiodic Markov chain with a nite or in nite state space We present a proof of convergence with probabili...

A new fully automated shape learning method is presented. It is based on clustering a set of training shapes in the original shape space (de ned by the coordinates of the contour points) and performing a Procrustes analysis on each cluster to obtain cluster prototypes and information about shape variation. The main di erence from previously reported methods is that the training set is rst autom...

the relationship between female oviposition preference and offspring performance is a key area of study on the evolution of host specificity. this study investigates host prefer-ence of the polyphagous hover fly syrphus ribesii (diptera: syrphidae) and compares it with some components of its offspring`s performance. females and offspring were tested on six naturally used aphid hosts (blackberry...

Direct numerical simulation of transitional and turbulent hypersonic boundary layers using the Navier-Stokes equations requires high-order accurate numerical methods to resolve a wide range of time and length scales. Compact or explicit nite di erence methods used for such simulation have mainly been central di erence schemes containing only phase errors without any numerical dissipation. Centr...

Recently, distributed shared memory systems have received much attention because such an abstraction simpliies programming. In this paper , we present a data consistency protocol for a distributed system which implements sequentially consistent memories. The protocol is aimed at an environment where no special support for atomic broadcast exists. As compared to previously proposed protocols, ou...

We discuss the temporal-di erence learning algorithm, as applied to approximating the cost-to-go function of an in nite-horizon discounted Markov chain. The algorithm we analyze updates parameters of a linear function approximator on{line, during a single endless trajectory of an irreducible aperiodic Markov chain with a nite or in nite state space. We present a proof of convergence (with proba...

We examine the stability properties of a class of LTV di"erence equations on an in&nite-dimensional state space that arise in backstepping designs for parabolic PDEs. The nominal system matrix of the di"erence equation has a special structure: all of its powers have entries that are −1, 0, or 1, and all of the eigenvalues of the matrix are on the unit circle. The di"erence equation is driven by...

In this paper, a new viewpoint, "individuality" and "sociality", is introduced for analyzing the multi-agent system's behavior. A hypothesis is set up that the both behavioral characters emerge to generate different actions from the other agent to increase each agent's bene t. "Individuality" is de ned as the di erence of actions between agents based on the di erence of its internal information...

With growing concerns about childhood obesity and the associated health risks, several countries are considering banning advertising directed at the children. In this paper we study the e¤ect of a ban on advertizing to children imposed in the Canadian province of Quebec in 1980 on fastfood purchase propensity and expenditures. Fastfood is not only one of the most highly advertised product categ...