Computer Science > Machine Learning
[Submitted on 28 Nov 2023 (v1), last revised 14 Feb 2024 (this version, v2)]
Title:Attentional Graph Neural Networks for Robust Massive Network Localization
View PDF HTML (experimental)Abstract:In recent years, Graph neural networks (GNNs) have emerged as a prominent tool for classification tasks in machine learning. However, their application in regression tasks remains underexplored. To tap the potential of GNNs in regression, this paper integrates GNNs with attention mechanism, a technique that revolutionized sequential learning tasks with its adaptability and robustness, to tackle a challenging nonlinear regression problem: network localization. We first introduce a novel network localization method based on graph convolutional network (GCN), which exhibits exceptional precision even under severe non-line-of-sight (NLOS) conditions, thereby diminishing the need for laborious offline calibration or NLOS identification. We further propose an attentional graph neural network (AGNN) model, aimed at improving the limited flexibility and mitigating the high sensitivity to the hyperparameter of the GCN-based method. The AGNN comprises two crucial modules, each designed with distinct attention architectures to address specific issues associated with the GCN-based method, rendering it more practical in real-world scenarios. Experimental results substantiate the efficacy of our proposed GCN-based method and AGNN model, as well as the enhancements of AGNN model. Additionally, we delve into the performance improvements of AGNN model by analyzing it from the perspectives of dynamic attention and computational complexity.
Submission history
From: Wenzhong Yan [view email][v1] Tue, 28 Nov 2023 15:05:13 UTC (2,442 KB)
[v2] Wed, 14 Feb 2024 11:21:32 UTC (583 KB)
Current browse context:
cs.LG
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.