Max-Margin Nonparametric Latent Feature Models for Link Prediction

Abstract

Link prediction is a fundamental task in statistical network analysis. Recent advances have been made on learning flexible nonparametric Bayesian latent feature models for link prediction. In this paper, we present a max-margin learning method for such nonparametric latent feature relational models. Our approach attempts to unite the ideas of max-margin learning and Bayesian nonparametrics to discover discriminative latent features for link prediction. It inherits the advances of nonparametric Bayesian methods to infer the unknown latent social dimension, while for discriminative link prediction, it adopts the max-margin learning principle by minimizing a hinge-loss using the linear expectation operator, without dealing with a highly nonlinear link likelihood function. For posterior inference, we develop an efficient stochastic variational inference algorithm under a truncated mean-field assumption. Our methods can scale up to large-scale real networks with millions of entities and tens of millions of positive links. We also provide a full Bayesian formulation, which can avoid tuning regularization hyper-parameters. Experimental results on a diverse range of real datasets demonstrate the benefits inherited from max-margin learning and Bayesian nonparametric inference.

arXiv comment14 pages, 8 figures

References in corpus (1)

• Modeling homophily and stochastic equivalence in symmetric relational data

Cited by in corpus (18)

• Learning over Knowledge-Base Embeddings for Recommendation
• Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendation
• Modeling Relation Paths for Representation Learning of Knowledge Bases
• Gibbs Max-margin Topic Models with Data Augmentation
• Self-Supervised Hyperboloid Representations from Logical Queries over Knowledge Graphs
• Max-Margin Nonparametric Latent Feature Models for Link Prediction
• AdaGCN: Adaboosting Graph Convolutional Networks into Deep Models
• Online Bayesian Passive-Aggressive Learning
• Big Learning with Bayesian Methods
• Constrained Bayesian Inference for Low Rank Multitask Learning
• Discriminative Nonparametric Latent Feature Relational Models with Data Augmentation
• Fast Sampling for Bayesian Max-Margin Models
• Building Memory with Concept Learning Capabilities from Large-scale Knowledge Base
• The Feature-First Block Model
• Discriminative Relational Topic Models
• SStaGCN: Simplified stacking based graph convolutional networks
• Log-Normal Matrix Completion for Large Scale Link Prediction
• A Constrained Matrix-Variate Gaussian Process for Transposable Data