Robust convex biclustering with a tuning-free method

TL;DR

This paper introduces a robust, tuning-free convex biclustering algorithm using Huber loss, which effectively handles heavy-tailed data and outperforms traditional methods in simulations and biomedical applications.

Contribution

The paper proposes a novel tuning-free method for robust convex biclustering with Huber loss, eliminating the need for parameter tuning and improving performance on heavy-tailed data.

Findings

Outperforms traditional biclustering methods in simulations with heavy-tailed noise.

Effectively handles heavy-tailed data in biomedical applications.

Provides an R package for practical implementation.

Abstract

Biclustering is widely used in different kinds of fields including gene information analysis, text mining, and recommendation system by effectively discovering the local correlation between samples and features. However, many biclustering algorithms will collapse when facing heavy-tailed data. In this paper, we propose a robust version of convex biclustering algorithm with Huber loss. Yet, the newly introduced robustification parameter brings an extra burden to selecting the optimal parameters. Therefore, we propose a tuning-free method for automatically selecting the optimal robustification parameter with high efficiency. The simulation study demonstrates the more fabulous performance of our proposed method than traditional biclustering methods when encountering heavy-tailed noise. A real-life biomedical application is also presented. The R package RcvxBiclustr is available at https://github.com/YifanChen3/RcvxBiclustr.