Neural Network Matrix Factorization

Abstract

Data often comes in the form of an array or matrix. Matrix factorization techniques attempt to recover missing or corrupted entries by assuming that the matrix can be written as the product of two low-rank matrices. In other words, matrix factorization approximates the entries of the matrix by a simple, fixed function---namely, the inner product---acting on the latent feature vectors for the corresponding row and column. Here we consider replacing the inner product by an arbitrary function that we learn from the data at the same time as we learn the latent feature vectors. In particular, we replace the inner product by a multi-layer feed-forward neural network, and learn by alternating between optimizing the network for fixed latent features, and optimizing the latent features for a fixed network. The resulting approach---which we call neural network matrix factorization or NNMF, for short---dominates standard low-rank techniques on a suite of benchmark but is dominated by some recent proposals that take advantage of the graph features. Given the vast range of architectures, activation functions, regularizers, and optimization techniques that could be used within the NNMF framework, it seems likely the true potential of the approach has yet to be reached.

Minor modifications to notation. Added additional experiments and discussion. 7 pages, 2 tables

References in corpus (3)

• A Review of Relational Machine Learning for Knowledge Graphs
• Leveraging Deep Neural Networks and Knowledge Graphs for Entity Disambiguation
• The Ladder: A Reliable Leaderboard for Machine Learning Competitions

Cited by in corpus (35)

• Deep Learning based Recommender System: A Survey and New Perspectives
• Graph Convolutional Matrix Completion
• Ask the GRU: Multi-Task Learning for Deep Text Recommendations
• A Neural Autoregressive Approach to Collaborative Filtering
• Solving Cold Start Problem in Recommendation with Attribute Graph Neural Networks
• Attacking Recommender Systems with Augmented User Profiles
• Deep Models of Interactions Across Sets
• STAR-GCN: Stacked and Reconstructed Graph Convolutional Networks for Recommender Systems
• On Universal Features for High-Dimensional Learning and Inference
• Metric Factorization: Recommendation beyond Matrix Factorization
• Inductive Matrix Completion Based on Graph Neural Networks
• Hybrid Collaborative Filtering with Autoencoders
• Why Not to Use Zero Imputation? Correcting Sparsity Bias in Training Neural Networks
• News Recommender System: A review of recent progress, challenges, and opportunities
• Neural Tensor Factorization
• Large-scale Collaborative Filtering with Product Embeddings
• Transfer Meets Hybrid: A Synthetic Approach for Cross-Domain Collaborative Filtering with Text
• DeepRec: An Open-source Toolkit for Deep Learning based Recommendation
• Deep Feature Factorization For Concept Discovery
• Spectral Geometric Matrix Completion
• Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback
• Extendable Neural Matrix Completion
• User Embedding based Neighborhood Aggregation Method for Inductive Recommendation
• A Biased Deep Tensor Factorization Network For Tensor Completion
• UserReg: A Simple but Strong Model for Rating Prediction
• Variational inference for neural network matrix factorization and its application to stochastic blockmodeling
• Quaternion Collaborative Filtering for Recommendation
• LCMR: Local and Centralized Memories for Collaborative Filtering with Unstructured Text
• Inductive Matrix Completion Using Graph Autoencoder
• A Deep, Forgetful Novelty-Seeking Movie Recommender Model
• Deep Unified Multimodal Embeddings for Understanding both Content and Users in Social Media Networks
• Collaborative Self-Attention for Recommender Systems
• Personalized Neural Embeddings for Collaborative Filtering with Text
• Scalable Recommender Systems through Recursive Evidence Chains
• On the Effectiveness of Low-rank Approximations for Collaborative Filtering compared to Neural Networks