Polynomial Networks and Factorization Machines: New Insights and Efficient Training Algorithms

Abstract

Polynomial networks and factorization machines are two recently-proposed models that can efficiently use feature interactions in classification and regression tasks. In this paper, we revisit both models from a unified perspective. Based on this new view, we study the properties of both models and propose new efficient training algorithms. Key to our approach is to cast parameter learning as a low-rank symmetric tensor estimation problem, which we solve by multi-convex optimization. We demonstrate our approach on regression and recommender system tasks.

Cited by in corpus (12)

• CnGAN: Generative Adversarial Networks for Cross-network user preference generation for non-overlapped users
• Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems
• Multimodal Polynomial Fusion for Detecting Driver Distraction
• Learning from Multi-View Multi-Way Data via Structural Factorization Machines
• Residual Tensor Train: A Quantum-inspired Approach for Learning Multiple Multilinear Correlations
• Detecting Beneficial Feature Interactions for Recommender Systems
• Feature Interaction-aware Graph Neural Networks
• The Second Order Linear Model
• Generalized Embedding Machines for Recommender Systems
• Enhancing Factorization Machines with Generalized Metric Learning
• Factorization Machines with Regularization for Sparse Feature Interactions
• Attribute-aware Collaborative Filtering: Survey and Classification