MetaWeighting paper overview
The paper In this blogpost we’ll expore the paper: MetaWeighting: Learning to Weight Tasks in Multi-Task Learning authored by: Yuren Mao, Zekai Wang, Weiwei Liu, Xuemin Lin and Pengtao Xie, The paper introduces a novel algorithm to dynamically adjust weights in Multi Task Learning setting via learning-to-learn paradigm. Multi-Task learning Multi-Task Learning (MTL) is a powerful concept in machine learning. It allows to share weights of the model between multiple tasks, it forces the model to learn a shared representation that can generalize better than if each task was learned in isolation. ...