Learning Latent Features with Pairwise Penalties in Matrix Completion

02/16/2018
by   Kaiyi Ji, et al.
0

Low-rank matrix completion (MC) has achieved great success in many real-world data applications. A latent feature model formulation is usually employed and, to improve prediction performance, the similarities between latent variables can be exploited by pairwise learning, e.g., the graph regularized matrix factorization (GRMF) method. However, existing GRMF approaches often use a squared L2 norm to measure the pairwise difference, which may be overly influenced by dissimilar pairs and lead to inferior prediction. To fully empower pairwise learning for matrix completion, we propose a general optimization framework that allows a rich class of (non-)convex pairwise penalty functions. A new and efficient algorithm is further developed to uniformly solve the optimization problem, with a theoretical convergence guarantee. In an important situation where the latent variables form a small number of subgroups, its statistical guarantee is also fully characterized. In particular, we theoretically characterize the complexity-regularized maximum likelihood estimator, as a special case of our framework. It has a better error bound when compared to the standard trace-norm regularized matrix completion. We conduct extensive experiments on both synthetic and real datasets to demonstrate the superior performance of this general framework.

READ FULL TEXT

page 14

page 27

research
09/13/2018

Clipped Matrix Completion: a Remedy for Ceiling Effects

We consider the recovery of a low-rank matrix from its clipped observati...
research
06/27/2017

Reexamining Low Rank Matrix Factorization for Trace Norm Regularization

Trace norm regularization is a widely used approach for learning low ran...
research
03/02/2013

Matrix Completion via Max-Norm Constrained Optimization

Matrix completion has been well studied under the uniform sampling model...
research
12/03/2015

Bayesian Matrix Completion via Adaptive Relaxed Spectral Regularization

Bayesian matrix completion has been studied based on a low-rank matrix f...
research
07/29/2020

A regularized deep matrix factorized model of matrix completion for image restoration

It has been an important approach of using matrix completion to perform ...
research
01/02/2022

Matrix Completion with Hierarchical Graph Side Information

We consider a matrix completion problem that exploits social or item sim...
research
12/10/2020

A generalised log-determinant regularizer for online semi-definite programming and its applications

We consider a variant of online semi-definite programming problem (OSDP)...

Please sign up or login with your details

Forgot password? Click here to reset