Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MPXGAT: An Attention based Deep Learning Model for Multiplex Graphs Embedding

Marco Bongiovanni, Luca Gallo, Roberto Grasso, Alfredo Pulvirenti

TL;DR

This work introduces MPXGAT, an innovative attention-based deep learning model tailored to multiplex graph embedding, which captures the structure of multiplex networks by harnessing both intra-layer and inter-layer connections.

Abstract

Graph representation learning has rapidly emerged as a pivotal field of study. Despite its growing popularity, the majority of research has been confined to embedding single-layer graphs, which fall short in representing complex systems with multifaceted relationships. To bridge this gap, we introduce MPXGAT, an innovative attention-based deep learning model tailored to multiplex graph embedding. Leveraging the robustness of Graph Attention Networks (GATs), MPXGAT captures the structure of multiplex networks by harnessing both intra-layer and inter-layer connections. This exploitation facilitates accurate link prediction within and across the network's multiple layers. Our comprehensive experimental evaluation, conducted on various benchmark datasets, confirms that MPXGAT consistently outperforms state-of-the-art competing algorithms.

MPXGAT: An Attention based Deep Learning Model for Multiplex Graphs Embedding

TL;DR

This work introduces MPXGAT, an innovative attention-based deep learning model tailored to multiplex graph embedding, which captures the structure of multiplex networks by harnessing both intra-layer and inter-layer connections.

Abstract

Graph representation learning has rapidly emerged as a pivotal field of study. Despite its growing popularity, the majority of research has been confined to embedding single-layer graphs, which fall short in representing complex systems with multifaceted relationships. To bridge this gap, we introduce MPXGAT, an innovative attention-based deep learning model tailored to multiplex graph embedding. Leveraging the robustness of Graph Attention Networks (GATs), MPXGAT captures the structure of multiplex networks by harnessing both intra-layer and inter-layer connections. This exploitation facilitates accurate link prediction within and across the network's multiple layers. Our comprehensive experimental evaluation, conducted on various benchmark datasets, confirms that MPXGAT consistently outperforms state-of-the-art competing algorithms.
Paper Structure (12 sections, 3 equations, 2 figures, 6 tables)

This paper contains 12 sections, 3 equations, 2 figures, 6 tables.

Table of Contents

  1. Introduction and Related Works
  2. Methods
  3. Preliminary Notions
  4. The MPXGAT General Framework
  5. MPXGAT Implementation for Link Prediction
  6. Results
  7. Dataset
  8. Experimental setup
  9. Competing Methods
  10. Embedding Multiplex Networks
  11. Measure the impact of Horizontal Embeddings
  12. Conclusions

Figures (2)

  • Figure 1: A toy example of a multiplex network with 2 horizontal layers. The solid edges represent the intra-layer connections while the dashed edges are the inter-layer edges.
  • Figure 2: The structure of the MPXGAT model. In this toy example, it is applied on a multiplex network with 2 horizontal layers where the solid blue edges represent the intra-layer connections while the dashed red edges are the inter-layer edges. The data is provided to the MPXGAT-H throughout the dot-and-dash blue lines. Once processed these are used to feed the MPXGAT-V together with the inter-layer links (the dotted green lines). The output of the model consists of both horizontal and vertical nodes embedding.