DeepAI AI Chat
Log In Sign Up

Co-manifold learning with missing data

by   Gal Mishne, et al.
NC State University
Yale University

Representation learning is typically applied to only one mode of a data matrix, either its rows or columns. Yet in many applications, there is an underlying geometry to both the rows and the columns. We propose utilizing this coupled structure to perform co-manifold learning: uncovering the underlying geometry of both the rows and the columns of a given matrix, where we focus on a missing data setting. Our unsupervised approach consists of three components. We first solve a family of optimization problems to estimate a complete matrix at multiple scales of smoothness. We then use this collection of smooth matrix estimates to compute pairwise distances on the rows and columns based on a new multi-scale metric that implicitly introduces a coupling between the rows and the columns. Finally, we construct row and column representations from these multi-scale metrics. We demonstrate that our approach outperforms competing methods in both data visualization and clustering.


page 15

page 16


A Sherman-Morrison-Woodbury Identity for Rank Augmenting Matrices with Application to Centering

Matrices of the form A + (V_1 + W_1)G(V_2 + W_2)^* are considered where ...

High-Rank Matrix Completion and Subspace Clustering with Missing Data

This paper considers the problem of completing a matrix with many missin...

Clustering, multicollinearity, and singular vectors

Let A be a matrix with its pseudo-matrix A^† and set S=I-A^†A. We prove ...

An Approximation Ratio for Biclustering

The problem of biclustering consists of the simultaneous clustering of r...

Projective Decomposition and Matrix Equivalence up to Scale

A data matrix may be seen simply as a means of organizing observations i...

Biclustering Using Modified Matrix Bandwidth Minimization and Biogeography-based Optimization

Data matrix having different sets of entities in its rows and columns ar...

Biclustering with Alternating K-Means

Biclustering is the task of simultaneously clustering the rows and colum...