Towards Compact and Explainable Deep Belief Networks
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F24%3A10490935" target="_blank" >RIV/00216208:11320/24:10490935 - isvavai.cz</a>
Result on the web
<a href="https://doi.org/10.1109/IJCNN60899.2024.10650547" target="_blank" >https://doi.org/10.1109/IJCNN60899.2024.10650547</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/IJCNN60899.2024.10650547" target="_blank" >10.1109/IJCNN60899.2024.10650547</a>
Alternative languages
Result language
angličtina
Original language name
Towards Compact and Explainable Deep Belief Networks
Original language description
The article investigates viable opportunities to extract knowledge from Deep Belief Networks and explain it understandably. The so-called confidence rules constitute an elegant means to express quantitative reasoning performed by the network. The paper introduces a new aproach to extracting confidence rules from the networks that allows for an enhanced accuracy of the inference. Compared to rival rule extraction techniques, supporting experiments confirm a significant improvement in inference accuracy obtained with less than one-fifth of the original network weights.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace
Others
Publication year
2024
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
IJCNN 2024 Conference Proceedings
ISBN
979-8-3503-5931-2
ISSN
2161-4393
e-ISSN
—
Number of pages
9
Pages from-to
11054-11062
Publisher name
IEEE
Place of publication
New York, NY
Event location
Yokohama
Event date
Jun 30, 2024
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—