A New Approach to Collaborative Filtering: Operator Estimation with Spectral Regularization

Abstract

We present a general approach for collaborative filtering (CF) using spectral regularization to learn linear operators from "users" to the "objects" they rate. Recent low-rank type matrix completion approaches to CF are shown to be special cases. However, unlike existing regularization based CF methods, our approach can be used to also incorporate information such as attributes of the users or the objects -- a limitation of existing regularization based CF methods. We then provide novel representer theorems that we use to develop new estimation methods. We provide learning algorithms based on low-rank decompositions, and test them on a standard CF dataset. The experiments indicate the advantages of generalizing the existing regularization based CF methods to incorporate related information about users and objects. Finally, we show that certain multi-task learning methods can be also seen as special cases of our proposed approach.

References in corpus (3)

• Empirical Analysis of Predictive Algorithms for Collaborative Filtering
• Consistency of trace norm minimization
• Low-rank matrix factorization with attributes

Cited by in corpus (36)

• Estimation of high-dimensional low-rank matrices
• Consistency of trace norm minimization
• Rank regularization and Bayesian inference for tensor completion and extrapolation
• Tight, robust, and feasible quantum speed limits for open dynamics
• In-network Sparsity-regularized Rank Minimization: Algorithms and Applications
• Operator-valued Kernels for Learning from Functional Response Data
• Learning Output Kernels for Multi-Task Problems
• Convergence Rates for Learning Linear Operators from Noisy Data
• Low-Rank Covariance Function Estimation for Multidimensional Functional Data
• Electricity Market Forecasting via Low-Rank Multi-Kernel Learning
• PAC-Bayesian Generalization Bound on Confusion Matrix for Multi-Class Classification
• Statistical analysis of $k$-nearest neighbor collaborative recommendation
• Learning Tensors in Reproducing Kernel Hilbert Spaces with Multilinear Spectral Penalties
• Sharp oracle inequalities for the prediction of a high-dimensional matrix
• Efficient Pairwise Learning Using Kernel Ridge Regression: an Exact Two-Step Method
• Stability of Multi-Task Kernel Regression Algorithms
• Structured Prediction for Conditional Meta-Learning
• Dyadic Prediction Using a Latent Feature Log-Linear Model
• Fast and Sample Efficient Inductive Matrix Completion via Multi-Phase Procrustes Flow
• Collaborative Filtering with Label Consistent Restricted Boltzmann Machine
• Link Prediction in Graphs with Autoregressive Features
• Flexible Low-Rank Statistical Modeling with Side Information
• The Advantage of Conditional Meta-Learning for Biased Regularization and Fine-Tuning
• Low-rank optimization methods based on projected projected-gradient descent that accumulate at Bouligand stationary points
• Vector-output ReLU Neural Network Problems are Copositive Programs: Convex Analysis of Two Layer Networks and Polynomial-time Algorithms
• Nonparametric Operator-Regularized Covariance Function Estimation for Functional Data
• A Stochastic Model for Collaborative Recommendation
• Non-parametric Structural Change Detection in Multivariate Systems
• Transformers are Deep Infinite-Dimensional Non-Mercer Binary Kernel Machines
• Matrix Completion via Nonconvex Regularization: Convergence of the Proximal Gradient Algorithm
• Structured low-rank matrix learning: algorithms and applications
• Learning Schatten--von Neumann Operators
• A Riemannian gossip approach to subspace learning on Grassmann manifold
• Data-driven modeling of beam loss in the LHC
• Low Rank Quaternion Matrix Recovery via Logarithmic Approximation
• An analysis of noise folding for low-rank matrix recovery