bibtype	C - Conference Paper (international conference)
ARLID	0490309
utime	20240103220122.3
mtime	20180614235959.9
title (primary) (eng)	Gradient Descent Parameter Learning of Bayesian Networks under Monotonicity Restrictions
specification	page_count 12 s. media_type P
serial	ARLID cav_un_epca*0490306 ISBN 978-80-7378-361-7 title Proceedings of the 11th Workshop on Uncertainty Processing (WUPES’18) page_num 153-164 publisher place Praha name MatfyzPress, Publishing House of the Faculty of Mathematics and Physics Charles University year 2018 editor name1 Kratochvíl name2 Václav editor name1 Vejnarová name2 Jiřina
keyword	Bayesian networks
keyword	Learning model parameters
keyword	monotonicity condition
author (primary)	ARLID cav_un_auth*0329423 name1 Plajner name2 Martin full_dept (cz) Matematická teorie rozhodování full_dept (eng) Department of Decision Making Theory department (cz) MTR department (eng) MTR institution UTIA-B full_dept Department of Decision Making Theory country CZ fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
author	ARLID cav_un_auth*0101228 name1 Vomlel name2 Jiří full_dept (cz) Matematická teorie rozhodování full_dept Department of Decision Making Theory department (cz) MTR department MTR institution UTIA-B full_dept Department of Decision Making Theory fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
source	url http://library.utia.cas.cz/separaty/2018/MTR/plajner-0490309.pdf
cas_special	project project_id GA16-12010S agency GA ČR country CZ ARLID cav_un_auth*0332303 project project_id SGS17/198/OHK4/3T/14 agency ČVUT ARLID cav_un_auth*0361640 country CZ abstract (eng) Learning parameters of a probabilistic model is a necessary step in most machine learning modeling tasks. When the model is complex and data volume is small the learning process may fail to provide good results. In this paper we present a method to improve learning results for small data sets by using additional information about the modelled system. This additional information is represented by monotonicity conditions which are restrictions on parameters of the model. Monotonicity simplifies the learning process and also these conditions are often required by the user of the system to hold. \n\nIn this paper we present a generalization of the previously used algorithm for parameter learning of Bayesian Networks under monotonicity conditions. This generalization allows both parents and children in the network to have multiple states. The algorithm is described in detail as well as monotonicity conditions are.\n\nThe presented algorithm is tested on two different data sets. Models are trained on differently sized data subsamples with the proposed method and the general EM algorithm. Learned models are then compared by their ability to fit data. We present empirical results showing the benefit of monotonicity conditions. The difference is especially significant when working with small data samples. The proposed method outperforms the EM algorithm for small sets and provides comparable results for larger sets. action ARLID cav_un_auth*0361637 name Workshop on Uncertainty Processing (WUPES’18) dates 20180606 place Třeboň country CZ mrcbC20-s 20180609 RIV JD FORD0 10000 FORD1 10200 FORD2 10201 reportyear 2019 num_of_auth 2 inst_support RVO:67985556 permalink http://hdl.handle.net/11104/0284592 confidential S arlyear 2018 mrcbU14 SCOPUS mrcbU24 PUBMED mrcbU34 WOS mrcbU63 cav_un_epca*0490306 Proceedings of the 11th Workshop on Uncertainty Processing (WUPES’18) MatfyzPress, Publishing House of the Faculty of Mathematics and Physics Charles University 2018 Praha 153 164 978-80-7378-361-7 mrcbU67 340 Kratochvíl Václav mrcbU67 340 Vejnarová Jiřina