AAAI 2023technical13 citations
Feature Decomposition for Reducing Negative Transfer: A Novel Multi-Task Learning Method for Recommender System (Student Abstract)
Jie Zhou, Qian Yu, Chuan Luo, Jing Zhang
Abstract
We propose a novel multi-task learning method termed Feature Decomposition Network (FDN). The key idea of the proposed FDN is to reduce the phenomenon of feature redundancy by explicitly decomposing features into task-specific features and task-shared features with carefully designed constraints. Experimental results show that our proposed FDN can outperform the state-of-the-art (SOTA) methods by a noticeable margin on Ali-CCP.
BibTeX
@article{Zhou_Yu_Luo_Zhang_2024, title={Feature Decomposition for Reducing Negative Transfer: A Novel Multi-Task Learning Method for Recommender System (Student Abstract)}, volume={37}, url={https://ojs.aaai.org/index.php/AAAI/article/view/27055}, DOI={10.1609/aaai.v37i13.27055}, abstractNote={We propose a novel multi-task learning method termed Feature Decomposition Network (FDN). The key idea of the proposed FDN is to reduce the phenomenon of feature redundancy by explicitly decomposing features into task-specific features and task-shared features with carefully designed constraints. Experimental results show that our proposed FDN can outperform the state-of-the-art (SOTA) methods by a noticeable margin on Ali-CCP.}, number={13}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Zhou, Jie and Yu, Qian and Luo, Chuan and Zhang, Jing}, year={2024}, month={Jul.}, pages={16390-16391} }