A Unified Generative Adversarial Learning Framework for Improvement of Skip-Gram Network Representation Learning Methods

Network Representation Learning (NRL), which aims to embed nodes into a latent, low-dimensional vector space while preserving some network properties, facilitates the further network analysis tasks. The goal of most NRL methods is to make similar nodes represented similarly in the embedding space. Many methods adopt the skip-gram model to achieve such goal by maximizing the predictive probability among the context nodes for each center node. The context nodes are usually determined based on the concept of proximity which is defined based on some explicit network features. However, these proximities may result in a loss of training samples and have limited discriminative power. We propose a general and unified generative adversarial learning framework to address the problems. The proposed framework can handle almost all kinds of networks in a unified way, including homogeneous plain networks, attribute augmented networks and heterogeneous networks. It can improve the performances of the most of the state-of-the-art skip-gram based NRL methods. Moreover, another unified and general NRL method is extended from the framework. It can learn the network representation independently. Extensive experiments on proximity preserving evaluation and two network analysis tasks, i.e., link prediction and node classifications, demonstrate the superiority and versatility of our framework.

Automated summary

Network Representation Learning (NRL) aims to embed nodes into a latent, low-dimensional vector space while preserving network properties. Many NRL methods use skip-gram model to achieve this by maximizing predictive probability among context nodes. However, these explicit network features may lead to loss of training samples and limited discriminative power. We propose a general and unified generative adversarial learning framework to address these issues and improve the performances of skip-gram based NRL methods.