RMTL

Regularized Multi-task Learning in R

Description

This package provides an efficient implementation of regularized multi-task learning comprising 10 algorithms applicable for regression, classification, joint feature selection, task clustering, low-rank learning, sparse learning and network incorporation. All algorithms are implemented basd on the accelerated gradient descent method and feature a complexity of O(1/k^2). Sparse model structure is induced by the solving the proximal operator. The package has been uploaded in the CRAN: https://CRAN.R-project.org/package=RMTL

Required Packages

Four packages have to be instaled in advanced to enable functions i.e. eigen-decomposition, 2D plotting: ‘MASS’, ‘psych’, ‘corpcor’ and ‘fields’. You can install them from the CRAN.

install.packages("MASS")
install.packages("psych")
install.packages("corpcor")
install.packages("fields")

Installation

You can choose any of the three ways to install RMTL.

  1. Install from CRAN in R environment (Recommend)
install.packages("RMTL")
# in this way, the requirement for installation are automatically checked.
  1. Install from github in R environment
install.packages("devtools")
library("devtools")
install_github("transbioZI/RMTL")
  1. Install from the source code
git clone https://github.com/transbioZI/RMTL.git
R CMD build ./RMTL/
R CMD INSTALL RMTL*.tar.gz

Tutorial

The tutorial of multi-task learning using RMTL can be found here.

Manual

Please check “RMTL-manuel.pdf” for more details.

Reference

Cao, Han, Jiayu Zhou and Emanuel Schwarz. “RMTL: An R Library for Multi-Task Learning.” Bioinformatics (2018).

Contact

If you have any question, please contact: [email protected]