A Bayesian Matrix Model for Relational Data
نویسندگان
چکیده
Relational learning can be used to augment one data source with other correlated sources of information, to improve predictive accuracy. We frame a large class of relational learning problems as matrix factorization problems, and propose a hierarchical Bayesian model. Training our Bayesian model using random-walk Metropolis-Hastings is impractically slow, and so we develop a block Metropolis-Hastings sampler which uses the gradient and Hessian of the likelihood to dynamically tune the proposal. We demonstrate that a predictive model of brain response to stimuli can be improved by augmenting it with side information about the stimuli.
منابع مشابه
An Efficient Bayesian Optimal Design for Logistic Model
Consider a Bayesian optimal design with many support points which poses the problem of collecting data with a few number of observations at each design point. Under such a scenario the asymptotic property of using Fisher information matrix for approximating the covariance matrix of posterior ML estimators might be doubtful. We suggest to use Bhattcharyya matrix in deriving the information matri...
متن کاملA Bayesian Matrix Factorization Model for Relational Data
Relational learning can be used to augment one data source with other correlated sources of information, to improve predictive accuracy. We frame a large class of relational learning problems as matrix factorization problems, and propose a hierarchical Bayesian model. Training our Bayesian model using random-walk Metropolis-Hastings is impractically slow, and so we develop a block MetropolisHas...
متن کاملConsistent Bayesian Network Scores for Multi-Relational Data
We describe a new method for extending, or upgrading, a Bayesian network (BN) score designed for i.i.d. data to multirelational data. The method defines a gain function that measures the difference in data fit between two structures. Our upgrade method preserves consistency, in the sense that if the i.i.d. score is consistent for i.i.d. data, the upgraded version is consistent for multi-relatio...
متن کاملInference Optimization using Relational Algebra
Exact inference procedures in Bayesian networks can be expressed using relational algebra; this provides a common ground for optimizations from the AI and database communities. Specifically, the ability to accomodate sparse representations of probability distributions opens up the way to optimize for their cardinality instead of the dimensionality ; we apply this in a sensor data model.
متن کاملA Grey-Based Fuzzy ELECTRE Model for Project Selection
Project selection is considered as an important problem in project management. It is multi-criteria in nature and is based on various quantitative and qualitative factors. The main purpose of this paper is to present a new rank-based method for project selection in outranking relation. According to this approach, decision alternatives were clustered in the concordance matrix and the discordance...
متن کامل