SP-GCRL: Influence Maximization on Incomplete Social Graphs

TL;DR

SP-GCRL is a reinforcement learning framework designed to optimize influence maximization on incomplete social graphs by modeling social propagation, learning robust node representations, and efficiently selecting seed nodes.

Contribution

It introduces a social-propagation-aware diffusion model, contrastive learning for robust representations, and an end-to-end seed selection policy using DDQN, addressing incomplete and noisy social graphs.

Findings

SP-GCRL outperforms heuristic and learning-based baselines across various networks.

It maintains strong scalability on large-scale networks.

The framework effectively handles incomplete and noisy social graphs.

Abstract

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.