Corpus ID: 121121054

Influence Maximization via Representation Learning

  title={Influence Maximization via Representation Learning},
  author={G. Panagopoulos and Michalis Vazirgiannis and Fragkiskos D. Malliaros},
Although influence maximization has been studied extensively in the past, the majority of works focus on the algorithmic aspect of the problem, overlooking several practical improvements that can be derived by data-driven observations or the inclusion of machine learning. The main challenges lie on the one hand on the computational demand of the algorithmic solution which restricts the scalability, and on the other the quality of the predicted influence spread. In this work, we propose… Expand
Multi-task Learning for Influence Estimation and Maximization
This work proposes IMINFECTOR (Influence Maximization with INFluencer vECTORs), a unified approach that uses representations learned from diffusion cascades to perform model-independent influence maximization that scales in real-world datasets. Expand
Learning policies for Social network discovery with Reinforcement learning
This work proposes a reinforcement learning framework for network discovery that automatically learns useful node and graph representations that encode important structural properties of the network. Expand


Model-Independent Online Learning for Influence Maximization
This work proposes a novel parametrization that not only makes the framework agnostic to the underlying diffusion model, but also statistically efficient to learn from data, and gives a corresponding monotone, submodular surrogate function. Expand
A Data-Based Approach to Social Influence Maximization
A new model is introduced, which is called credit distribution, that directly leverages available propagation traces to learn how influence flows in the network and uses this to estimate expected influence spread, and is time-aware in the sense that it takes the temporal nature of influence into account. Expand
DiffuGreedy: An Influence Maximization Algorithm Based on Diffusion Cascades
This work introduces a simple yet effective algorithm that combines the algorithmic methodology with the diffusion cascades and compares it with four different prevalent influence maximization approaches, on a large scale Chinese microblogging dataset. Expand
Sketch-based Influence Maximization and Computation: Scaling up with Guarantees
This work develops a novel sketch-based design for influence computation, called SKIM, which scales to graphs with billions of edges, with one to two orders of magnitude speedup over the best greedy methods. Expand
Inf2vec: Latent Representation Model for Social Influence Embedding
A new model Inf2vec is developed, which combines both the local influence neighborhood and global user similarity to learn the representations of nodes in a low-dimensional space and significantly outperforms state-of-the-art baseline algorithms. Expand
node2vec: Scalable Feature Learning for Networks
In node2vec, an algorithmic framework for learning continuous feature representations for nodes in networks, a flexible notion of a node's network neighborhood is defined and a biased random walk procedure is designed, which efficiently explores diverse neighborhoods. Expand
Learning influence probabilities in social networks
This paper proposes models and algorithms for learning the model parameters and for testing the learned models to make predictions, and develops techniques for predicting the time by which a user may be expected to perform an action. Expand
Learning Diffusion using Hyperparameters
This paper studies a natural restriction of the hypothesis class using additional information available in order to dramatically reduce the sample complexity of the learning process in the independent cascade (IC) model. Expand
DeepInf : Modeling Influence Locality in Large Social Networks
Online communities such as Facebook, Twitter, WeChat, andWeibo have become an indispensable part of our everyday life, where we can easily access the behaviors of our friends and are influenced byExpand
Multi-scale Information Diffusion Prediction with Reinforced Recurrent Networks
This paper proposes a novel multi-scale diffusion prediction model based on reinforcement learning (RL), which incorporates the macroscopic diffusion size information into the RNN-based microscopic diffusion model by addressing the non-differentiable problem. Expand