A Brief Review of Deep Multi-task Learning and Auxiliary Task Learning

TL;DR

This paper reviews recent deep multi-task learning approaches, focusing on how auxiliary tasks can be effectively selected to enhance model performance across multiple tasks.

Contribution

It provides a concise overview of recent dMTL methods and discusses strategies for selecting useful auxiliary tasks to improve main task performance.

Findings

Summarizes recent advances in deep multi-task learning

Highlights methods for auxiliary task selection

Discusses performance improvements through auxiliary tasks

Abstract

Multi-task learning (MTL) optimizes several learning tasks simultaneously and leverages their shared information to improve generalization and the prediction of the model for each task. Auxiliary tasks can be added to the main task to ultimately boost the performance. In this paper, we provide a brief review on the recent deep multi-task learning (dMTL) approaches followed by methods on selecting useful auxiliary tasks that can be used in dMTL to improve the performance of the model for the main task.