Next Article in Journal
The Odds Exponential-Pareto IV Distribution: Regression Model and Application
Next Article in Special Issue
Probabilistic Shaping for Finite Blocklengths: Distribution Matching and Sphere Shaping
Previous Article in Journal
Aspects of Chemical Entropy Generation in Flow of Casson Nanofluid between Radiative Stretching Disks
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Entropy
Volume 22
Issue 5
10.3390/e22050496
entropy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

On Training Neural Network Decoders of Rate Compatible Polar Codes via Transfer Learning

by
Hyunjae Lee
1,
Eun Young Seo
2,
Hyosang Ju
1 and
Sang-Hyo Kim
1,*
1
Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon 16419, Korea
2
Samsung Electronics, Hwaseong 18448, Korea
*
Author to whom correspondence should be addressed.
Entropy 2020, 22(5), 496; https://doi.org/10.3390/e22050496
Submission received: 3 March 2020 / Revised: 19 April 2020 / Accepted: 22 April 2020 / Published: 25 April 2020
(This article belongs to the Special Issue Information Theory for Communication Systems)
Browse Figures
Versions Notes

Abstract

Neural network decoders (NNDs) for rate-compatible polar codes are studied in this paper. We consider a family of rate-compatible polar codes which are constructed from a single polar coding sequence as defined by 5G new radios. We propose a transfer learning technique for training multiple NNDs of the rate-compatible polar codes utilizing their inclusion property. The trained NND for a low rate code is taken as the initial state of NND training for the next smallest rate code. The proposed method provides quicker training as compared to separate learning of the NNDs according to numerical results. We additionally show that an underfitting problem of NND training due to low model complexity can be solved by transfer learning techniques.
Keywords: polar codes; deep learning; neural network decoder; transfer learning polar codes; deep learning; neural network decoder; transfer learning

Share and Cite

MDPI and ACS Style

Lee, H.; Seo, E.Y.; Ju, H.; Kim, S.-H. On Training Neural Network Decoders of Rate Compatible Polar Codes via Transfer Learning. Entropy 2020, 22, 496. https://doi.org/10.3390/e22050496

AMA Style

Lee H, Seo EY, Ju H, Kim S-H. On Training Neural Network Decoders of Rate Compatible Polar Codes via Transfer Learning. Entropy. 2020; 22(5):496. https://doi.org/10.3390/e22050496

Chicago/Turabian Style

Lee, Hyunjae, Eun Young Seo, Hyosang Ju, and Sang-Hyo Kim. 2020. "On Training Neural Network Decoders of Rate Compatible Polar Codes via Transfer Learning" Entropy 22, no. 5: 496. https://doi.org/10.3390/e22050496

APA Style

Lee, H., Seo, E. Y., Ju, H., & Kim, S.-H. (2020). On Training Neural Network Decoders of Rate Compatible Polar Codes via Transfer Learning. Entropy, 22(5), 496. https://doi.org/10.3390/e22050496

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop