SP-GCRL: Influence Maximization on Incomplete Social Graphs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Niu, Haohua, Yang, Yuxuan, Zhang, Lingfeng, Li, Hao, Liang, Jiao, Luo, Zongfu, Rossi, Luca
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918498247114752
author Niu, Haohua
Yang, Yuxuan
Zhang, Lingfeng
Li, Hao
Liang, Jiao
Luo, Zongfu
Rossi, Luca
author_facet Niu, Haohua
Yang, Yuxuan
Zhang, Lingfeng
Li, Hao
Liang, Jiao
Luo, Zongfu
Rossi, Luca
contents Influence maximization (IM) in real platforms is challenged by incomplete, noisy social graphs and non-stationary diffusion dynamics. We propose SP-GCRL, a social-propagation-aware graph contrastive reinforcement learning framework that learns end-to-end seed selection under partial observability.We first introduce a social-propagation-aware nonlinear diffusion function to model reinforcement/diminishing effects and probability drift under repeated exposure; we then construct dual structural views and perform contrastive learning to obtain node representations robust to missing edges and weak ties, while replacing expensive strategy metrics with a GAT-based regression surrogate to improve efficiency and scalability; finally, we use DDQN to learn an end-to-end seed selection policy on top of these representations. Experiments on multiple real-world networks show that SP-GCRL achieves significant gains over heuristic and learning-based baselines across budgets and topologies, while maintaining strong large-scale scalability.
format Preprint
id arxiv_https___arxiv_org_abs_2605_12513
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SP-GCRL: Influence Maximization on Incomplete Social Graphs
Niu, Haohua
Yang, Yuxuan
Zhang, Lingfeng
Li, Hao
Liang, Jiao
Luo, Zongfu
Rossi, Luca
Social and Information Networks
Artificial Intelligence
Influence maximization (IM) in real platforms is challenged by incomplete, noisy social graphs and non-stationary diffusion dynamics. We propose SP-GCRL, a social-propagation-aware graph contrastive reinforcement learning framework that learns end-to-end seed selection under partial observability.We first introduce a social-propagation-aware nonlinear diffusion function to model reinforcement/diminishing effects and probability drift under repeated exposure; we then construct dual structural views and perform contrastive learning to obtain node representations robust to missing edges and weak ties, while replacing expensive strategy metrics with a GAT-based regression surrogate to improve efficiency and scalability; finally, we use DDQN to learn an end-to-end seed selection policy on top of these representations. Experiments on multiple real-world networks show that SP-GCRL achieves significant gains over heuristic and learning-based baselines across budgets and topologies, while maintaining strong large-scale scalability.
title SP-GCRL: Influence Maximization on Incomplete Social Graphs
topic Social and Information Networks
Artificial Intelligence
url https://arxiv.org/abs/2605.12513