Fetching the paper…

Pseudo-task Augmentation: From Deep Multitask Learning to Intratask Sharing---and Back · Around