Results 1 
9 of
9
LikelihoodBased Robust Classification with Bayesian Networks
"... Abstract. Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we ..."
Abstract

Cited by 4 (3 self)
 Add to MetaCart
(Show Context)
Abstract. Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihoodbased learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness.
A pairwise label ranking method with imprecise scores and partial predictions
 Machine Learning and Knowledge Discovery in Databases
, 2013
"... Abstract. In this paper, we are interested in the label ranking problem. We are more specifically interested in the recent trend consisting in predicting partial but more accurate (i.e., making less incorrect statements) orders rather than complete ones. To do so, we propose a ranking method based o ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
(Show Context)
Abstract. In this paper, we are interested in the label ranking problem. We are more specifically interested in the recent trend consisting in predicting partial but more accurate (i.e., making less incorrect statements) orders rather than complete ones. To do so, we propose a ranking method based on pairwise imprecise scores obtained from likelihood functions. We discuss how such imprecise scores can be aggregated to produce interval orders, which are specific types of partial orders. We then analyse the performances of the method as well as its sensitivity to missing data and parameter values.
†Department of Statistics, LMU Munich
"... Abstract. Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, w ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract. Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihoodbased learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness.
LikelihoodBased Robust Classification with Bayesian Networks
, 2012
"... Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a ..."
Abstract
 Add to MetaCart
Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihoodbased learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness. 1
Credal Ensembles of Classifiers
"... It is studied how to aggregate the probabilistic predictions generated by different SPODE (SuperParentOneDependence Estimators) classifiers. It is shown that aggregating such predictions via compressionbased weights achieves a slight but consistent improvement of performance over previously exis ..."
Abstract
 Add to MetaCart
(Show Context)
It is studied how to aggregate the probabilistic predictions generated by different SPODE (SuperParentOneDependence Estimators) classifiers. It is shown that aggregating such predictions via compressionbased weights achieves a slight but consistent improvement of performance over previously existing aggregation methods, including Bayesian Model Averaging and simple average (the approach adopted by the AODE algorithm). Then, attention is given to the problem of choosing the prior probability distribution over the models; this is an important issue in any Bayesian ensemble of models. To robustly deal with the choice of the prior, the single prior over the models is substituted by a set of priors over the models (credal set), thus obtaining a credal ensemble of Bayesian classifiers. The credal ensemble recognizes the priordependent instances, namely the instances whose most probable class varies when different prior over the models are considered. When faced with priordependent instances, the credal ensemble remains reliable by returning a set of classes rather than a single class. Two credal ensembles of SPODEs are developed; the first generalizes the Bayesian Model Averaging and the second the compressionbased aggregation. Extensive experiments show that the novel ensembles compare favorably to traditional methods for aggregating SPODEs and also to previous credal classifiers.
LikelihoodBased Robust Classification with Bayesian Networks
, 2012
"... Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a ..."
Abstract
 Add to MetaCart
Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihoodbased learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness. 1
Compressionbased AODE Classifiers
"... Abstract. We propose the COMPAODE classifier, which adopts the compressionbased approach [1] to average the posterior probabilities computed by different nonnaive classifiers (SPODEs). COMPAODE improves classification performance over the wellknown AODE [10] model. COMPAODE assumes a uniform pr ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract. We propose the COMPAODE classifier, which adopts the compressionbased approach [1] to average the posterior probabilities computed by different nonnaive classifiers (SPODEs). COMPAODE improves classification performance over the wellknown AODE [10] model. COMPAODE assumes a uniform prior over the SPODEs; we then develop the credal classifier COMPAODE*, substituting the uniform prior by a set of priors. COMPAODE * returns more classes when the classification is priordependent, namely if the most probable class varies with the prior adopted over the SPODEs. COMPAODE * achieves higher classification utility than both COMPAODE and AODE. 1
LikelihoodBased Robust Classification with Bayesian Networks
, 2012
"... Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a ..."
Abstract
 Add to MetaCart
Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihoodbased learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness. 1
Approximate Credal Network Updating by Linear Programming with Applications to Decision MakingI
"... Credal nets are probabilistic graphical models which extend Bayesian nets to cope with sets of distributions. An algorithm for approximate credal network updating is presented. The problem in its general formulation is a multilinear optimization task, which can be linearized by an appropriate rule f ..."
Abstract
 Add to MetaCart
Credal nets are probabilistic graphical models which extend Bayesian nets to cope with sets of distributions. An algorithm for approximate credal network updating is presented. The problem in its general formulation is a multilinear optimization task, which can be linearized by an appropriate rule for fixing all the local models apart from those of a single variable. This simple idea can be iterated and quickly leads to accurate inferences. A transformation is also derived to reduce decision making in credal networks based on the maximality criterion to updating. The decision task is proved to have the same complexity of standard inference, being NPPPcomplete for general credal nets and NPcomplete for polytrees. Similar results are derived for the Eadmissibility criterion. Numerical experiments confirm a good performance of the method.