Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.20585 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915759009038336 |
|---|---|
| author | Hao, Aiming Zhu, Chen Zhu, Jiashu Wu, Jiahong Chu, Xiangxiang |
| author_facet | Hao, Aiming Zhu, Chen Zhu, Jiashu Wu, Jiahong Chu, Xiangxiang |
| contents | Ordinal regression and ranking are challenging due to inherent ordinal dependencies that conventional methods struggle to model. We propose Ranking-Aware Reinforcement Learning (RARL), a novel RL framework that explicitly learns these relationships. At its core, RARL features a unified objective that synergistically integrates regression and Learning-to-Rank (L2R), enabling mutual improvement between the two tasks. This is driven by a ranking-aware verifiable reward that jointly assesses regression precision and ranking accuracy, facilitating direct model updates via policy optimization. To further enhance training, we introduce Response Mutation Operations (RMO), which inject controlled noise to improve exploration and prevent stagnation at saddle points. The effectiveness of RARL is validated through extensive experiments on three distinct benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20585 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Ranking-aware Reinforcement Learning for Ordinal Ranking Hao, Aiming Zhu, Chen Zhu, Jiashu Wu, Jiahong Chu, Xiangxiang Machine Learning Artificial Intelligence Ordinal regression and ranking are challenging due to inherent ordinal dependencies that conventional methods struggle to model. We propose Ranking-Aware Reinforcement Learning (RARL), a novel RL framework that explicitly learns these relationships. At its core, RARL features a unified objective that synergistically integrates regression and Learning-to-Rank (L2R), enabling mutual improvement between the two tasks. This is driven by a ranking-aware verifiable reward that jointly assesses regression precision and ranking accuracy, facilitating direct model updates via policy optimization. To further enhance training, we introduce Response Mutation Operations (RMO), which inject controlled noise to improve exploration and prevent stagnation at saddle points. The effectiveness of RARL is validated through extensive experiments on three distinct benchmarks. |
| title | Ranking-aware Reinforcement Learning for Ordinal Ranking |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2601.20585 |