The “Big Data” revolution is spawning systems designed to make decisions from data. In particular, deep learning methods have emerged as the state of the art method in many important breakthroughs [18, 20, 28]. This is due to the statistical flexibility and computational scalability of large and deep neural networks which allows them to harness the information of large and rich datasets. At the...

Most studies of Bayesian updating use experimental data. This paper uses a non-experimental data source– the voter ballots of the Associated Press (AP) college football poll, a weekly subjective ranking of the top 25 teams–to test Bayes’ rule as a descriptive model. I estimate the voters’ Bayesian posterior rankings for the first seven weeks of the 2006-08 seasons (a total of over 31,000 rankin...

We describe a Bayesian Reasoning Framework (BRF) that supports business rule operations for on-line information systems. BRF comprises a three-layer environment with business information systems at the top, a middle-ware Bayesian reasoning server, and a Bayesian reasoning engine at the bottom. The top and middle-ware layers communicate via SOAP/XML protocol, while the middle-ware and bottom lay...

Association rules discovered through attribute-oriented induction are commonly used in data mining tools to express relationships between variables. However, causal inference algorithms discover more concise relationships between variables, namely, relations of direct cause. These algorithms produce regressive structured equation models for continuous linear data and Bayes networks for discrete...

By combining the advantages of argument map and Bayesian network, a case description model based on evidence (CDMBE), which is suitable to continental law system, is proposed to describe the criminal cases. The logic of the model adopts the credibility logical reason and gets evidence-based reasoning quantitatively based on evidences. In order to consist with practical inference rules, five typ...

We study the classic problem of choosing a prior distribution for a location parameter β = (β1, . . . , βp) as p grows large. First, we study the standard “global-local shrinkage” approach, based on scale mixtures of normals. Two theorems are presented which characterize certain desirable properties of shrinkage priors for sparse problems. Next, we review some recent results showing how Lévy pr...

This paper compares the performance of simple voting rules, called referenda, to the performance of interim efficient mechanisms for the provision of a public good. In a referendum, voters simply vote for or against the provision of the public good, and production of the public good depends on whether or not the number of yes votes exceeds a prespecified threshold. Costs are shared equally. We ...