arrow
返回

Classification-Driven Discrete Neural Representation Learning for Semantic Communications

delete2024-05-01
delete1
PRE
AI
W
Wenhui Hua
L
Longhui Xiong
S
Sicong Liu
L
Lingyu Chen
X
Xuemin Hong *
J
João F. C. Mota
程
程翔 (Xiang Cheng)
DOI:10.1109/JIOT.2024.3354312delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Semantic communications is a key enabler of the Internet of Things (IoT). By focusing on the semantic meaning of data rather than bit-level recovery, it allows intelligent agents to communicate necessary information at much lower rates. A promising technique for semantic communications is discrete neural representation learning (DNRL). The main idea is to learn discrete symbols from low-level, high-dimensional sensory data, such that each symbol is grounded to a meaningful pattern in the sensory domain. This article proposes a DNRL scheme that integrates three mechanisms into a coherent framework: 1) contrastive learning; 2) sparse coding; and 3) neural index quantization. The proposed scheme is applied to public image data sets for lossy image compression with a downstream classification task. Results show that the proposed approach produces a highly compact continuous latent representation and a semantic discrete representation, with marginal degradation to the classification accuracy. The interpretability and consistency of the learned subsymbolic discrete representations are validated by experiments of neural-net dissection, neural-net visualization, and MaxAmp-K classification test, a concept that we propose to evaluate classification performance of extremely compressed signals. Finally, the discrete representations are shown to be useful in rate-adaptive distributed sensing applications at the low-to-medium signal-to-noise ratios (SNRs).
Keyword:
Data compression
distributed detection
image classification
image representations
neural networks
quantization

期刊

IEEE Internet of Things Journal 封面图
IEEE Internet of Things Journal
IF:
8.9
论文数:
1.4W
被引数:
7.8W

机构

H
Heriot Watt University
学者数:
6.0K
论文数: 6.5K
被引数: 57
P
peking university
学者数:
11.9W
论文数: 8.7W
被引数: 146
X
xiamen university
学者数:
5.9W
论文数: 3.8W
被引数: 67
学者 查看更多机构
引用论文

引用论文

An analysis of four different methods of producing focal cerebral ischemia with endothelin-1 in the rat
err2006-10-01
err0
PREAI
errV WINDLE; A SZYMANSKA; S GRANTERBUTTON; C WHITE; R BUIST; J PEELING; D CORBETT
err分享
err收藏
Effect of Inhibin on Testosterone Metabolism by Rat Ventral Prostate in Vitro
err2009-07-09
err0
PREAI
errR. Joseph; A. Maitra; S. B. Moodbidri; A. R. Sheth
err分享
err收藏
err分享
err收藏
err分享
err收藏
Pilot study of DNA methylation in the pathogenesis of chronic rhinosinusitis with nasal polyps
err2015-12-01
err0
errOAAI
errY.B. Zheng; Y. Zhao; L.Y. Yue; P. Lin; Y.F. Liu; J.M. Xian; G.Y. Zhou; D.Y. Wang
err分享
err收藏
err
IF0
err
err0
PREAI
err
err分享
err收藏
学者 查看更多内容