Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic

Sejin Seo; Jihong Park; Seung Woo Ko; Jinho Choi; Mehdi Bennis; Seong Lyun Kim

doi:10.1109/SECON58729.2023.10287426

Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic

Sejin Seo, Jihong Park, Seung Woo Ko, Jinho Choi, Mehdi Bennis, Seong Lyun Kim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we suggest a method to transform a communication protocol based on deep neural network (NN) into a semantic communication protocol. We need such transformation to alleviate the issues posed by NN's lack of interpretability and redundant parameters due to overparametrization. However, transformation process is challenging because it is difficult to disambiguate the semantics while reducing the protocol's complexity. We solve the challenge by employing NN's activation patterns and probabilistic logic. Lastly, we validate our method by transforming an NN trained for a medium access control (MAC) protocol and verifying its contention performance compared to ALOHA based protocols.

Original language	English
Title of host publication	2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023
Publisher	IEEE Computer Society
Pages	76-78
Number of pages	3
ISBN (Electronic)	9798350300529
DOIs	https://doi.org/10.1109/SECON58729.2023.10287426
State	Published - 2023
Event	20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023 - Madrid, Spain Duration: 11 Sep 2023 → 14 Sep 2023

Publication series

Name	Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops
Volume	2023-September
ISSN (Print)	2155-5486
ISSN (Electronic)	2155-5494

Conference

Conference	20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023
Country/Territory	Spain
City	Madrid
Period	11/09/23 → 14/09/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

MARL
medium access control (MAC)
protocol learning
semantic information theory
Semantic protocol

Access to Document

10.1109/SECON58729.2023.10287426

Cite this

Seo, S., Park, J., Ko, S. W., Choi, J., Bennis, M., & Kim, S. L. (2023). Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic. In 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023 (pp. 76-78). (Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops; Vol. 2023-September). IEEE Computer Society. https://doi.org/10.1109/SECON58729.2023.10287426

Seo, Sejin ; Park, Jihong ; Ko, Seung Woo et al. / Semantic Communication Protocol : Demystifying Deep Neural Networks via Probabilistic Logic. 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023. IEEE Computer Society, 2023. pp. 76-78 (Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops).

@inproceedings{c9eb86b75e204af192653d66e9c1d97a,

title = "Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic",

abstract = "In this paper, we suggest a method to transform a communication protocol based on deep neural network (NN) into a semantic communication protocol. We need such transformation to alleviate the issues posed by NN's lack of interpretability and redundant parameters due to overparametrization. However, transformation process is challenging because it is difficult to disambiguate the semantics while reducing the protocol's complexity. We solve the challenge by employing NN's activation patterns and probabilistic logic. Lastly, we validate our method by transforming an NN trained for a medium access control (MAC) protocol and verifying its contention performance compared to ALOHA based protocols.",

keywords = "MARL, medium access control (MAC), protocol learning, semantic information theory, Semantic protocol",

author = "Sejin Seo and Jihong Park and Ko, {Seung Woo} and Jinho Choi and Mehdi Bennis and Kim, {Seong Lyun}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023 ; Conference date: 11-09-2023 Through 14-09-2023",

year = "2023",

doi = "10.1109/SECON58729.2023.10287426",

language = "English",

series = "Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops",

publisher = "IEEE Computer Society",

pages = "76--78",

booktitle = "2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023",

address = "United States",

}

Seo, S, Park, J, Ko, SW, Choi, J, Bennis, M & Kim, SL 2023, Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic. in 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023. Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops, vol. 2023-September, IEEE Computer Society, pp. 76-78, 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023, Madrid, Spain, 11/09/23. https://doi.org/10.1109/SECON58729.2023.10287426

Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic. / Seo, Sejin; Park, Jihong; Ko, Seung Woo et al.
2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023. IEEE Computer Society, 2023. p. 76-78 (Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops; Vol. 2023-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Semantic Communication Protocol

T2 - 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023

AU - Seo, Sejin

AU - Park, Jihong

AU - Ko, Seung Woo

AU - Choi, Jinho

AU - Bennis, Mehdi

AU - Kim, Seong Lyun

PY - 2023

Y1 - 2023

N2 - In this paper, we suggest a method to transform a communication protocol based on deep neural network (NN) into a semantic communication protocol. We need such transformation to alleviate the issues posed by NN's lack of interpretability and redundant parameters due to overparametrization. However, transformation process is challenging because it is difficult to disambiguate the semantics while reducing the protocol's complexity. We solve the challenge by employing NN's activation patterns and probabilistic logic. Lastly, we validate our method by transforming an NN trained for a medium access control (MAC) protocol and verifying its contention performance compared to ALOHA based protocols.

AB - In this paper, we suggest a method to transform a communication protocol based on deep neural network (NN) into a semantic communication protocol. We need such transformation to alleviate the issues posed by NN's lack of interpretability and redundant parameters due to overparametrization. However, transformation process is challenging because it is difficult to disambiguate the semantics while reducing the protocol's complexity. We solve the challenge by employing NN's activation patterns and probabilistic logic. Lastly, we validate our method by transforming an NN trained for a medium access control (MAC) protocol and verifying its contention performance compared to ALOHA based protocols.

KW - MARL

KW - medium access control (MAC)

KW - protocol learning

KW - semantic information theory

KW - Semantic protocol

UR - http://www.scopus.com/inward/record.url?scp=85177449745&partnerID=8YFLogxK

U2 - 10.1109/SECON58729.2023.10287426

DO - 10.1109/SECON58729.2023.10287426

M3 - Conference contribution

AN - SCOPUS:85177449745

T3 - Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops

SP - 76

EP - 78

BT - 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023

PB - IEEE Computer Society

Y2 - 11 September 2023 through 14 September 2023

ER -

Seo S, Park J, Ko SW, Choi J, Bennis M, Kim SL. Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic. In 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2023. IEEE Computer Society. 2023. p. 76-78. (Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks workshops). doi: 10.1109/SECON58729.2023.10287426

Semantic Communication Protocol: Demystifying Deep Neural Networks via Probabilistic Logic

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this