MetaBalance: Improving Multi-Task Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks

Abstract

In many personalized recommendation scenarios, the generalization ability of a target task can be improved via learning with additional auxiliary tasks alongside this target task on a multi-task network. However, this method often suffers from a serious optimization imbalance problem. On the one hand, one or more auxiliary tasks might have a larger influence than the target task and even dominate the network weights, resulting in worse recommendation accuracy for the target task. On the other hand, the influence of one or more auxiliary tasks might be too weak to assist the target task. More challenging is that this imbalance dynamically changes throughout the training process and varies across the parts of the same network. We propose a new method: MetaBalance to balance auxiliary losses via directly manipulating their gradients w.r.t the shared parameters in the multi-task network. Specifically, in each training iteration and adaptively for each part of the network, the gradient of an auxiliary loss is carefully reduced or enlarged to have a closer magnitude to the gradient of the target loss, preventing auxiliary tasks from being so strong that dominate the target task or too weak to help the target task. Moreover, the proximity between the gradient magnitudes can be flexibly adjusted to adapt MetaBalance to different scenarios. The experiments show that our proposed method achieves a significant improvement of 8.34% in terms of NDCG@10 upon the strongest baseline on two real-world datasets. The code of our approach can be found at here: https://github.com/facebookresearch/MetaBalance

arXiv commentAccepted by the WebConf 2022

References in corpus (13)

• Adam: A Method for Stochastic Optimization
• An Overview of Multi-Task Learning in Deep Neural Networks
• Multi-Task Learning for Dense Prediction Tasks: A Survey
• GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks
• Deep Learning Recommendation Model for Personalization and Recommendation Systems
• Ask the GRU: Multi-Task Learning for Deep Text Recommendations
• Auxiliary Tasks in Multi-task Learning
• Self-Supervised Generalisation with Meta Auxiliary Learning
• Adapting Auxiliary Losses Using Gradient Similarity
• Learning Longer-term Dependencies in RNNs with Auxiliary Losses
• Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout
• Entire Space Multi-Task Model: An Effective Approach for Estimating Post-Click Conversion Rate
• A Hierarchical Self-Attentive Model for Recommending User-Generated Item Lists

Cited by in corpus (7)

• Multi-behavior Self-supervised Learning for Recommendation
• AdaTask: A Task-aware Adaptive Learning Rate Approach to Multi-task Learning
• Remote Sensing SpatioTemporal Vision-Language Models: A Comprehensive Survey
• M3oE: Multi-Domain Multi-Task Mixture-of Experts Recommendation Framework
• GradCraft: Elevating Multi-task Recommendations through Holistic Gradient Crafting
• Unconstrained Monotonic Calibration of Predictions in Deep Ranking Systems
• ORCA: Mitigating Over-Reliance for Multi-Task Dwell Time Prediction with Causal Decoupling