$\alpha$VIL: Learning to Leverage Auxiliary Tasks for Multitask Learning

arXiv:2405.07769v1 Announce Type: new
Abstract: Multitask Learning is a Machine Learning paradigm that aims to train a range of (usually related) tasks with the help of a shared model. While the goal is often to improve the joint performance of all training tasks, another approach is to focus on the performance of a specific target task, while treating the remaining ones as auxiliary data from which to possibly leverage positive transfer towards the target during training. In such settings, it …

abstract alpha arxiv cs.lg focus machine machine learning multitask learning paradigm performance tasks train training type while