A General MCMC Method for Bayesian Inference in Logic-Based Probabilistic Modeling
نویسنده
چکیده
We propose a general MCMC method for Bayesian inference in logic-based probabilistic modeling. It covers a broad class of generative models including Bayesian networks and PCFGs. The idea is to generalize an MCMC method for PCFGs to the one for a Turing-complete probabilistic modeling language PRISM in the context of statistical abduction where parse trees are replaced with explanations. We describe how to estimate the marginal probability of data from MCMC samples and how to perform Bayesian Viterbi inference using an example of Naive Bayes model augmentedwith a hidden variable.
منابع مشابه
Knowledge-leveraged Computational Thinking through Natural Language Processing and Statistical Logic (NII Shonan Meeting 2011-4)
This talk describes a recent effort on the development of a textual entailment data set. Rather than assuming a sub-component of applications like question answering and multi-document summarization, we focus on a realworld task to judge whether a natural language proposition is true or false according to a given text. I will describe the design of resource development and features of the obtai...
متن کاملBayesian inference for statistical abduction using Markov chain Monte Carlo
Abduction is one of the basic logical inferences (deduction, induction and abduction) and derives the best explanations for our observation. Statistical abduction attempts to define a probability distribution over explanations and to evaluate them by their probabilities. The framework of statistical abduction is general since many well-known probabilistic models, i.e., BNs, HMMs and PCFGs, are ...
متن کاملA Probabilistic Knowledge Base Using Annotated Bayesian Network Features
The probabilistic modeling of a high dimensional domain includes the modeling of the joint distribution over the domain variables on numeric, qualitative and possibly causal levels. Additionally, it includes the combination of statistical data with domain knowledge acquired from experts and the usage of the result in a decision theoretic framework. We overview the Bayesian network representatio...
متن کاملGeneral-Purpose MCMC Inference over Relational Structures
Tasks such as record linkage and multi-target tracking, which involve reconstructing the set of objects that underlie some observed data, are particularly challenging for probabilistic inference. Recent work has achieved efficient and accurate inference on such problems using Markov chain Monte Carlo (MCMC) techniques with customized proposal distributions. Currently, implementing such a system...
متن کاملRisk Analysis of Operating Room Using the Fuzzy Bayesian Network Model
To enhance Patient’s safety, we need effective methods for risk management. This work aims to propose an integrated approach to risk management for a hospital system. To improve patient’s safety, we should develop flexible methods where different aspects of risk and type of information are taken into consideration. This paper proposes a fuzzy Bayesian network to model and analyze risk in the op...
متن کامل