Structural extension to logistic regression: Discriminative parameter learning of belief net classifiers

DMCA

Structural extension to logistic regression: Discriminative parameter learning of belief net classifiers (2002)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Russell Greiner , Wei Zhou , Xiaoyuan Su , Bin Shen

Venue:	In Proceedings of the Eighteenth Annual National Conference on Artificial Intelligence (AAAI-02
Citations:	76 - 8 self

BibTeX

@INPROCEEDINGS{Greiner02structuralextension,
    author = {Russell Greiner and Wei Zhou and Xiaoyuan Su and Bin Shen},
    title = {Structural extension to logistic regression: Discriminative parameter learning of belief net classifiers},
    booktitle = {In Proceedings of the Eighteenth Annual National Conference on Artificial Intelligence (AAAI-02},
    year = {2002},
    pages = {167--173}
}

OpenURL

Abstract

Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to find the BN that maximizes a different objective function — viz., likelihood, rather than classification accuracy — typically by first learning an appropriate graphical structure, then finding the maximal likelihood parameters for that structure. As these parameters may not maximize the classification accuracy, “discriminative learners” follow the alternative approach of seeking the parameters that maximize conditional likelihood (CL), over the distribution of instances the BN will have to classify. This paper first formally specifies this task, and shows how it extends standard logistic regression. After analyzing its inherent sample and computational complexity, we present a general algorithm for this task, ELR, which applies to arbitrary BN structures and which works effectively even when given incomplete training data. This paper presents empirical evidence thatELR produces better classifiers than are produced by the standard “generative” algorithms in a variety of situations, especially in common situations where the given BN-structure is incorrect. Keywords: (Bayesian) belief nets, Logistic regression, Classification, PAC-learning, Computational/sample complexity 1

Keyphrases

logistic regression structural extension belief net classifier discriminative parameter learning classification accuracy likely class label incomplete training data common situation maximal likelihood parameter empirical evidence thatelr appropriate graphical structure different objective function viz computational sample complexity computational complexity bayesian belief net discriminative learner classification task inherent sample standard generative algorithm many bn-learners bn structure standard logistic regression general algorithm conditional likelihood belief net alternative approach

DMCA