التفاصيل البيبلوغرافية
| العنوان: |
Multitask learning with single gradient step update for task balancing. |
| المؤلفون: |
Lee, Sungjae1 (AUTHOR) sungjaelee@dgu.ac.kr, Son, Youngdoo1 (AUTHOR) youngdoo@dongguk.edu |
| المصدر: |
Neurocomputing. Jan2022, Vol. 467, p442-453. 12p. |
| مصطلحات موضوعية: |
*COMPUTER vision, *COMPUTER multitasking, *CONVOLUTIONAL neural networks, *MACHINE learning, *ALGORITHMS |
| مستخلص: |
• We propose a novel multitask learning algorithm to alleviate the imbalance problem. • The proposed method is inspired by gradient-based meta-learning. • It consists of the single gradient step update and alternating training procedure. • It outperforms the state-of-the-art methods in several computer vision problems. Multitask learning is a methodology to boost generalization performance and also reduce computational intensity and memory usage. However, learning multiple tasks simultaneously can be more difficult than learning a single task because it can cause imbalance among tasks. To address the imbalance problem, we propose an algorithm to balance between tasks at the gradient level by applying gradient-based meta -learning to multitask learning. The proposed method trains shared layers and task-specific layers separately so that the two layers with different roles in a multitask network can be fitted to their own purposes. In particular, the shared layer that contains informative knowledge shared among tasks is trained by employing single gradient step update and inner/outer loop training to mitigate the imbalance problem at the gradient level. We apply the proposed method to various multitask computer vision problems and achieve state-of-the-art performance. [ABSTRACT FROM AUTHOR] |
| قاعدة البيانات: |
Academic Search Index |
