Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout

Abstract

The vast majority of deep models use multiple gradient signals, typically corresponding to a sum of multiple loss terms, to update a shared set of trainable weights. However, these multiple updates can impede optimal training by pulling the model in conflicting directions. We present Gradient Sign Dropout (GradDrop), a probabilistic masking procedure which samples gradients at an activation layer based on their level of consistency. GradDrop is implemented as a simple deep layer that can be used in any deep net and synergizes with other gradient balancing approaches. We show that GradDrop outperforms the state-of-the-art multiloss methods within traditional multitask and transfer learning settings, and we discuss how GradDrop reveals links between optimal multiloss training and gradient stochasticity.

arXiv commentConference on Neural Information Processing Systems (NeurIPS) 2020

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Sequence to Sequence Learning with Neural Networks
• ADADELTA: An Adaptive Learning Rate Method
• An overview of gradient descent optimization algorithms
• SGDR: Stochastic Gradient Descent with Warm Restarts
• Multi-Task Learning for Dense Prediction Tasks: A Survey
• The Loss Surfaces of Multilayer Networks
• Don't Decay the Learning Rate, Increase the Batch Size
• On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
• Multi-Task Learning as Multi-Objective Optimization
• One Model To Learn Them All
• Multi-Task Deep Neural Networks for Natural Language Understanding
• Unsupervised Domain Adaptation through Self-Supervision
• Gradient Surgery for Multi-Task Learning
• Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks
• Pseudo-task Augmentation: From Deep Multitask Learning to Intratask Sharing---and Back
• Regularizing Meta-Learning via Gradient Dropout

Cited by in corpus (3)

• AdaTask: A Task-aware Adaptive Learning Rate Approach to Multi-task Learning
• MetaBalance: Improving Multi-Task Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks
• T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler