LGMLMay 23, 2018

Matrix Co-completion for Multi-label Classification with Missing Features and Labels

arXiv:1805.09156v117 citations
Originality Incremental advance
AI Analysis

This work addresses a challenging multi-label classification scenario with missing data, offering a novel approach that could benefit applications in domains like text or image tagging, though it appears incremental by building on matrix completion techniques.

The paper tackles the problem of multi-label classification with missing features and labels by proposing a co-completion algorithm that assumes a latent matrix with low-rank structure, addressing limitations of existing linear methods and achieving improved performance in experiments.

We consider a challenging multi-label classification problem where both feature matrix $\X$ and label matrix $\Y$ have missing entries. An existing method concatenated $\X$ and $\Y$ as $[\X; \Y]$ and applied a matrix completion (MC) method to fill the missing entries, under the assumption that $[\X; \Y]$ is of low-rank. However, since entries of $\Y$ take binary values in the multi-label setting, it is unlikely that $\Y$ is of low-rank. Moreover, such assumption implies a linear relationship between $\X$ and $\Y$ which may not hold in practice. In this paper, we consider a latent matrix $\Z$ that produces the probability $σ(Z_{ij})$ of generating label $Y_{ij}$, where $σ(\cdot)$ is nonlinear. Considering label correlation, we assume $[\X; \Z]$ is of low-rank, and propose an MC algorithm based on subgradient descent named co-completion (COCO) motivated by elastic net and one-bit MC. We give a theoretical bound on the recovery effect of COCO and demonstrate its practical usefulness through experiments.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes