Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916497523212288 |
|---|---|
| author | Zhang, Zhi Shen, Jiayi Cao, Congfeng Dai, Gaole Zhou, Shiji Zhang, Qizhe Zhang, Shanghang Shutova, Ekaterina |
| author_facet | Zhang, Zhi Shen, Jiayi Cao, Congfeng Dai, Gaole Zhou, Shiji Zhang, Qizhe Zhang, Shanghang Shutova, Ekaterina |
| contents | Advancing towards generalist agents necessitates the concurrent processing of multiple tasks using a unified model, thereby underscoring the growing significance of simultaneous model training on multiple downstream tasks. A common issue in multi-task learning is the occurrence of gradient conflict, which leads to potential competition among different tasks during joint training. This competition often results in improvements in one task at the expense of deterioration in another. Although several optimization methods have been developed to address this issue by manipulating task gradients for better task balancing, they cannot decrease the incidence of gradient conflict. In this paper, we systematically investigate the occurrence of gradient conflict across different methods and propose a strategy to reduce such conflicts through sparse training (ST), wherein only a portion of the model's parameters are updated during training while keeping the rest unchanged. Our extensive experiments demonstrate that ST effectively mitigates conflicting gradients and leads to superior performance. Furthermore, ST can be easily integrated with gradient manipulation techniques, thus enhancing their effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_18615 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective Zhang, Zhi Shen, Jiayi Cao, Congfeng Dai, Gaole Zhou, Shiji Zhang, Qizhe Zhang, Shanghang Shutova, Ekaterina Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Advancing towards generalist agents necessitates the concurrent processing of multiple tasks using a unified model, thereby underscoring the growing significance of simultaneous model training on multiple downstream tasks. A common issue in multi-task learning is the occurrence of gradient conflict, which leads to potential competition among different tasks during joint training. This competition often results in improvements in one task at the expense of deterioration in another. Although several optimization methods have been developed to address this issue by manipulating task gradients for better task balancing, they cannot decrease the incidence of gradient conflict. In this paper, we systematically investigate the occurrence of gradient conflict across different methods and propose a strategy to reduce such conflicts through sparse training (ST), wherein only a portion of the model's parameters are updated during training while keeping the rest unchanged. Our extensive experiments demonstrate that ST effectively mitigates conflicting gradients and leads to superior performance. Furthermore, ST can be easily integrated with gradient manipulation techniques, thus enhancing their effectiveness. |
| title | Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2411.18615 |