Adaptive polynomial filters with individual learning rates for computationally efficient lung tumor motion prediction
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21220%2F15%3A00236641" target="_blank" >RIV/68407700:21220/15:00236641 - isvavai.cz</a>
Result on the web
<a href="http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7347077" target="_blank" >http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7347077</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/IWCIM.2015.7347077" target="_blank" >10.1109/IWCIM.2015.7347077</a>
Alternative languages
Result language
angličtina
Original language name
Adaptive polynomial filters with individual learning rates for computationally efficient lung tumor motion prediction
Original language description
This paper presents a study of higher-order neural units as polynomial adaptive filters with multiple-learning-rate gradient descent for 3-D lung tumor motion prediction. The method is compared with single-learning rate gradient descent approaches with and without learning rate normalization. Experimental analysis is done with linear and quadratic neural unit. The influence of correct selection of adaptation parameters and the dependence of learning time on accuracy were experimentally analyzed. The prediction accuracy is nearly equal to recently published results of batch retraining approaches while the computational efficiency is higher for the introduced approach.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
BC - Theory and management systems
OECD FORD branch
—
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2015
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
INTERNATIONAL WORKSHOP ON COMPUTATIONAL INTELLIGENCE FOR MULTIMEDIA UNDERSTANDING
ISBN
978-1-4673-8457-5
ISSN
—
e-ISSN
—
Number of pages
5
Pages from-to
—
Publisher name
České vysoké učení technické v Praze
Place of publication
Praha
Event location
Praha
Event date
Oct 29, 2015
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000380431200017