Belief Propagation and Bethe approximation for Traffic Prediction

نویسندگان

  • Cyril Furtlehner
  • Arnaud de La Fortelle
  • Jean-Marc Lasgouttes
چکیده

We define and study an inference algorithm based on“belief propagation” (BP) and the Bethe approximation. The idea is to encode into a graph an a priori information composed of correlations or marginal probabilities of variables, and to use a message passing procedure to estimate the actual state from some extra realtime information. This method is originally designed for traffic prediction and is particularly suitable in settings where the only information available is floating car data. We propose a discretized traffic description, based on the Ising model of statistical physics, in order to both reconstruct and predict the traffic in real time. General properties of BP are addressed in this context. In particular, a detailed study of stability is proposed with respect to the a priori data and the graph topology. The behavior of the algorithm is illustrated by numerical studies on a simple traffic toy model. How this approach can be generalized to encode superposition of many traffic patterns is discussed. Key-words: belief propagation algorithm, Bethe approximation, traffic prediction, intelligent transport systems, floating car data ∗ INRIA Futurs – LRI, Bat. 490, Université Paris-Sud – 91405 Orsay cedex (France) † INRIA Rocquencourt – Domaine de Voluceau B.P. 105 – 78153 Le Chesnay cedex (France) ‡ École des Mines de Paris – CAOR research centre – 60, boulevard Saint-Michel – 75272 Paris cedex 06 (France) Propagation de croyances et approximation de Bethe pour la prédiction de trafic Résumé : On définit et étudie un algorithme de reconstruction utilisant l’algorithme « Belief Propagation » (propagation de croyances, BP) et l’approximation de Bethe. L’idée est d’encoder dans un graphe des données a priori composées de corrélations ou de lois marginales et d’utiliser une procédure de passage de messages pour estimer l’état réel à partir d’informations temps-réel. Cette méthode, développée pour des besoins de prédiction de trafic, est particulièrement adaptée au cas où la seule information disponible provient de véhicules sonde (Floating Car Data). Nous proposons une discrétisation binaire du trafic s’appuyant sur le modèle d’Ising de physique statistique, permettant de reconstruire et de prédire le trafic en temps réel. Des propriétés générales de l’algorithme BP sont discutées dans ce contexte. En particulier une étude détaillée des propriétés de stabilité fonction des données a priori et de la topologie du graphe est fournie. Une étude numérique sur un modèle de trafic simplifié permet d’illustrer le fonctionnement de l’algorithme. La façon de généraliser cette approche pour encoder une superposition de plusieurs états de trafic est discutée. Mots-clés : propagation de croyances, approximation de Bethe, reconstruction de trafic, prédiction, systèmes de transport intelligent, véhicules traceurs Belief Propagation and Bethe approximation for Traffic Prediction 3

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Bethe and Related Pairwise Entropy Approximations

For undirected graphical models, belief propagation often performs remarkably well for approximate marginal inference, and may be viewed as a heuristic to minimize the Bethe free energy. Focusing on binary pairwise models, we demonstrate that several recent results on the Bethe approximation may be generalized to a broad family of related pairwise free energy approximations with arbitrary count...

متن کامل

New Generalizations of the Bethe Approximation via Asymptotic Expansion

The Bethe approximation, discovered in statistical physics, gives an efficient algorithm called belief propagation (BP) for approximating a partition function. BP empirically gives an accurate approximation for many problems, e.g., low-density parity-check codes, compressed sensing, etc. Recently, Vontobel gives a novel characterization of the Bethe approximation using graph cover. In this pape...

متن کامل

Understanding the Bethe Approximation: When and How can it go Wrong?

Belief propagation is a remarkably effective tool for inference, even when applied to networks with cycles. It may be viewed as a way to seek the minimum of the Bethe free energy, though with no convergence guarantee in general. A variational perspective shows that, compared to exact inference, this minimization employs two forms of approximation: (i) the true entropy is approximated by the Bet...

متن کامل

Understanding the Bethe approximation: when and how does it go wrong?

Belief propagation is a remarkably effective tool for inference, even when applied to networks with cycles. It may be viewed as a way to seek the minimum of the Bethe free energy, though with no convergence guarantee in general. A variational perspective shows that, compared to exact inference, this minimization employs two forms of approximation: (i) the true entropy is approximated by the Bet...

متن کامل

New Understanding of the Bethe Approximation and the Replica Method

In this thesis, new generalizations of the Bethe approximation and new understanding of the replica method are proposed. The Bethe approximation is an efficient approximation for graphical models, which gives an asymptotically accurate estimate of the partition function for many graphical models. The Bethe approximation explains the well-known message passing algorithm, belief propagation, whic...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007