Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.01665 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911390773542912 |
|---|---|
| author | Li, Wensheng Tian, Yichao Wang, Hao Zhou, Ruifeng Guan, Hanting Zhang, Chao Tao, Dacheng |
| author_facet | Li, Wensheng Tian, Yichao Wang, Hao Zhou, Ruifeng Guan, Hanting Zhang, Chao Tao, Dacheng |
| contents | Curriculum learning (CL) mimics human learning, in which easy samples are learned first, followed by harder samples, and has become an effective method for training deep networks. However, many existing automatic CL methods maintain a preference for easy samples during the entire training process regardless of the constantly evolving training state. This is just like a human curriculum that fails to provide individualized instruction, which can delay learning progress. To address this issue, we propose an adaptively point-weighting (APW) curriculum learning method that assigns a weight to each training sample based on its training loss. The weighting strategy of APW follows the easy-to-hard training paradigm, guided by the current training state of the network. We present a theoretical analysis of APW, including training effectiveness, training stability, and generalization performance. Experimental results validate these theoretical findings and demonstrate the superiority of the proposed APW method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_01665 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adaptively Point-weighting Curriculum Learning Li, Wensheng Tian, Yichao Wang, Hao Zhou, Ruifeng Guan, Hanting Zhang, Chao Tao, Dacheng Machine Learning Curriculum learning (CL) mimics human learning, in which easy samples are learned first, followed by harder samples, and has become an effective method for training deep networks. However, many existing automatic CL methods maintain a preference for easy samples during the entire training process regardless of the constantly evolving training state. This is just like a human curriculum that fails to provide individualized instruction, which can delay learning progress. To address this issue, we propose an adaptively point-weighting (APW) curriculum learning method that assigns a weight to each training sample based on its training loss. The weighting strategy of APW follows the easy-to-hard training paradigm, guided by the current training state of the network. We present a theoretical analysis of APW, including training effectiveness, training stability, and generalization performance. Experimental results validate these theoretical findings and demonstrate the superiority of the proposed APW method. |
| title | Adaptively Point-weighting Curriculum Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2505.01665 |