Graph Influence Maximization Algorithm Based on Reinforcement Learning-PPO Algorithm

  • Wenxin Zhang
  • , Yaofei Ma*
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Information dissemination in social network has gained great attention in the study of complex networks. In this paper, we propose a novel network influence maximization algorithm incorporating graph representation learning and reinforcement learning, called RL-PPO, to select a set of seed nodes as the starter nodes and by which the information can be propagated as fast as could in the network,i.e., reaching the status of influence maximization. This algorithm take the feature matrix and adjacency matrix of the network as inputs, and the seed nodes as outputs.To better represent the network, the graph representation learning method is employed to extract network features and to perform feature aggregation. The experiment results show that, compared with the traditional greedy algorithm, the seed nodes determined by this proposed algorithm GraphPPO exhibited higher propagation efficiency.

Original languageEnglish
Title of host publicationProceedings of 2022 Chinese Intelligent Systems Conference - Volume I
EditorsYingmin Jia, Weicun Zhang, Yongling Fu, Shoujun Zhao
PublisherSpringer Science and Business Media Deutschland GmbH
Pages563-572
Number of pages10
ISBN (Print)9789811962028
DOIs
StatePublished - 2022
Event18th Chinese Intelligent Systems Conference, CISC 2022 - Beijing, China
Duration: 15 Oct 202216 Oct 2022

Publication series

NameLecture Notes in Electrical Engineering
Volume950 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

Conference18th Chinese Intelligent Systems Conference, CISC 2022
Country/TerritoryChina
CityBeijing
Period15/10/2216/10/22

Keywords

  • Complex networks
  • Graph representation learning
  • Influence maximization
  • Reinforcement learning

Fingerprint

Dive into the research topics of 'Graph Influence Maximization Algorithm Based on Reinforcement Learning-PPO Algorithm'. Together they form a unique fingerprint.

Cite this