Dealing with soft evidence in credal networks

نویسندگان

  • José Carlos F. da Rocha
  • Alaine M. Guimarães
  • Cassio P. de Campos
چکیده

Abstract. The formalism of credal networks can be used to represent imprecision in multivariate probabilistic models. Currently, the algorithms for inference with credal nets allow to compute posterior probability intervals given specific evidence. However, they do not deal with soft evidence. This paper presents an approach to integrate soft evidence in credal networks. The proposal is to convert the soft evidence into constraints that are appended to a multilinear program to perform inferences.

برای دانلود رایگان 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

Integração de evidências em redes credais e a regra de Jeffrey

Credal networks provide a scheme for dealing with imprecise probabilistic models. The inference algorithms often used in credal networks compute the interval of the posterior probability of an event of interest given evidence of the specific kind evidence that describe the current state of a set of variables. These algorithms do not perform evidential reasoning in case of the evidence must be p...

متن کامل

Reliable Uncertain Evidence Modeling in Bayesian Networks by Credal Networks

A reliable modeling of uncertain evidence in Bayesian networks based on a set-valued quantification is proposed. Both soft and virtual evidences are considered. We show that evidence propagation in this setup can be reduced to standard updating in an augmented credal network, equivalent to a set of consistent Bayesian networks. A characterization of the computational complexity for this task is...

متن کامل

Credal Compositional Models and Credal Networks

This paper studies the composition operator for credal sets introduced at the last ISIPTA conference in more detail. Our main attention is devoted to the relationship between a special type of compositional model, so-called perfect sequences of credal sets, and those of (precise) probability distributions, with the goal of finding the relationship between credal compositional models and credal ...

متن کامل

Locally specified credal networks

Credal networks are models that extend Bayesian nets to deal with imprecision in probability, and can actually be regarded as sets of Bayesian nets. Evidence suggests that credal nets are a powerful means to represent and deal with many important and challenging problems in uncertain reasoning. We give examples to show that some of these problems can only be modelled by credal nets called non-s...

متن کامل

Credal Sets Approximation by Lower Probabilities: Application to Credal Networks

Credal sets are closed convex sets of probability mass functions. The lower probabilities specified by a credal set for each element of the power set can be used as constraints defining a second credal set. This simple procedure produces an outer approximation, with a bounded number of extreme points, for general credal sets. The approximation is optimal in the sense that no other lower probabi...

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

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2012