Informative relationship multi-task learning: Exploring pairwise contribution across tasks' sharing knowledge

Xiangchao Chang, Menghui Zhou, Xulong Wang, Yun Yang, Po Yang

Abstract: Highlights•An informative framework to quantify mutual knowledge contribution while ensuring diversity.•A sparse regularization to handle the shared knowledge while enhancing efficacy.•Accelerated gradient descent-based method to address point-wise regularization.