The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dalam makalah ini, isu pengiraan dalam masalah pembelajaran rangkaian kepercayaan Bayesian (BBN) berdasarkan prinsip panjang penerangan minimum (MDL) ditangani. Berdasarkan formula asimptotik panjang perihalan, kami menggunakan teknik cawangan dan terikat untuk mencari struktur rangkaian sebenar. Carian algoritma yang terhasil banyak menjimatkan pengiraan namun berjaya mencari struktur rangkaian dengan nilai minimum formula. Setakat ini, belum ada algoritma carian yang mencari penyelesaian optimum untuk contoh saiz praktikal dan satu set struktur rangkaian dalam erti kata kebarangkalian posterior maksimum, dan carian heuristik seperti perangkap K2 dan K3 dalam optima tempatan disebabkan oleh ketamakan. alam semula jadi walaupun saiz sampel adalah besar. Algoritma yang dicadangkan, kerana ia meminimumkan panjang perihalan, akhirnya memilih struktur rangkaian sebenar apabila saiz sampel pergi ke infiniti.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Joe SUZUKI, "Learning Bayesian Belief Networks Based on the MDL Principle: An Efficient Algorithm Using the Branch and Bound Technique" in IEICE TRANSACTIONS on Information,
vol. E82-D, no. 2, pp. 356-367, February 1999, doi: .
Abstract: In this paper, the computational issue in the problem of learning Bayesian belief networks (BBNs) based on the minimum description length (MDL) principle is addressed. Based on an asymptotic formula of description length, we apply the branch and bound technique to finding true network structures. The resulting algorithm searches considerably saves the computation yet successfully searches the network structure with the minimum value of the formula. Thus far, there has been no search algorithm that finds the optimal solution for examples of practical size and a set of network structures in the sense of the maximum posterior probability, and heuristic searches such as K2 and K3 trap in local optima due to the greedy nature even when the sample size is large. The proposed algorithm, since it minimizes the description length, eventually selects the true network structure as the sample size goes to infinity.
URL: https://global.ieice.org/en_transactions/information/10.1587/e82-d_2_356/_p
Salinan
@ARTICLE{e82-d_2_356,
author={Joe SUZUKI, },
journal={IEICE TRANSACTIONS on Information},
title={Learning Bayesian Belief Networks Based on the MDL Principle: An Efficient Algorithm Using the Branch and Bound Technique},
year={1999},
volume={E82-D},
number={2},
pages={356-367},
abstract={In this paper, the computational issue in the problem of learning Bayesian belief networks (BBNs) based on the minimum description length (MDL) principle is addressed. Based on an asymptotic formula of description length, we apply the branch and bound technique to finding true network structures. The resulting algorithm searches considerably saves the computation yet successfully searches the network structure with the minimum value of the formula. Thus far, there has been no search algorithm that finds the optimal solution for examples of practical size and a set of network structures in the sense of the maximum posterior probability, and heuristic searches such as K2 and K3 trap in local optima due to the greedy nature even when the sample size is large. The proposed algorithm, since it minimizes the description length, eventually selects the true network structure as the sample size goes to infinity.},
keywords={},
doi={},
ISSN={},
month={February},}
Salinan
TY - JOUR
TI - Learning Bayesian Belief Networks Based on the MDL Principle: An Efficient Algorithm Using the Branch and Bound Technique
T2 - IEICE TRANSACTIONS on Information
SP - 356
EP - 367
AU - Joe SUZUKI
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E82-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 1999
AB - In this paper, the computational issue in the problem of learning Bayesian belief networks (BBNs) based on the minimum description length (MDL) principle is addressed. Based on an asymptotic formula of description length, we apply the branch and bound technique to finding true network structures. The resulting algorithm searches considerably saves the computation yet successfully searches the network structure with the minimum value of the formula. Thus far, there has been no search algorithm that finds the optimal solution for examples of practical size and a set of network structures in the sense of the maximum posterior probability, and heuristic searches such as K2 and K3 trap in local optima due to the greedy nature even when the sample size is large. The proposed algorithm, since it minimizes the description length, eventually selects the true network structure as the sample size goes to infinity.
ER -