PCA-based Multi Task Learning: a Random Matrix Approach
Machine Learning
2021-11-02 v1 Machine Learning
Statistics Theory
Statistics Theory
Abstract
The article proposes and theoretically analyses a \emph{computationally efficient} multi-task learning (MTL) extension of popular principal component analysis (PCA)-based supervised learning schemes \cite{barshan2011supervised,bair2006prediction}. The analysis reveals that (i) by default learning may dramatically fail by suffering from \emph{negative transfer}, but that (ii) simple counter-measures on data labels avert negative transfer and necessarily result in improved performances. Supporting experiments on synthetic and real data benchmarks show that the proposed method achieves comparable performance with state-of-the-art MTL methods but at a \emph{significantly reduced computational cost}.
Cite
@article{arxiv.2111.00924,
title = {PCA-based Multi Task Learning: a Random Matrix Approach},
author = {Malik Tiomoko and Romain Couillet and Frédéric Pascal},
journal= {arXiv preprint arXiv:2111.00924},
year = {2021}
}