Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.15729 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918124627951616 |
|---|---|
| author | Montreuil, Yannis Yeo, Shu Heng Carlier, Axel Ng, Lai Xing Ooi, Wei Tsang |
| author_facet | Montreuil, Yannis Yeo, Shu Heng Carlier, Axel Ng, Lai Xing Ooi, Wei Tsang |
| contents | The Two-Stage Learning-to-Defer (L2D) framework has been extensively studied for classification and, more recently, regression tasks. However, many real-world applications require solving both tasks jointly in a multi-task setting. We introduce a novel Two-Stage L2D framework for multi-task learning that integrates classification and regression through a unified deferral mechanism. Our method leverages a two-stage surrogate loss family, which we prove to be both Bayes-consistent and $(\mathcal{G}, \mathcal{R})$-consistent, ensuring convergence to the Bayes-optimal rejector. We derive explicit consistency bounds tied to the cross-entropy surrogate and the $L_1$-norm of agent-specific costs, and extend minimizability gap analysis to the multi-expert two-stage regime. We also make explicit how shared representation learning -- commonly used in multi-task models -- affects these consistency guarantees. Experiments on object detection and electronic health record analysis demonstrate the effectiveness of our approach and highlight the limitations of existing L2D methods in multi-task scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_15729 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Two-Stage Learning-to-Defer Approach for Multi-Task Learning Montreuil, Yannis Yeo, Shu Heng Carlier, Axel Ng, Lai Xing Ooi, Wei Tsang Machine Learning Human-Computer Interaction The Two-Stage Learning-to-Defer (L2D) framework has been extensively studied for classification and, more recently, regression tasks. However, many real-world applications require solving both tasks jointly in a multi-task setting. We introduce a novel Two-Stage L2D framework for multi-task learning that integrates classification and regression through a unified deferral mechanism. Our method leverages a two-stage surrogate loss family, which we prove to be both Bayes-consistent and $(\mathcal{G}, \mathcal{R})$-consistent, ensuring convergence to the Bayes-optimal rejector. We derive explicit consistency bounds tied to the cross-entropy surrogate and the $L_1$-norm of agent-specific costs, and extend minimizability gap analysis to the multi-expert two-stage regime. We also make explicit how shared representation learning -- commonly used in multi-task models -- affects these consistency guarantees. Experiments on object detection and electronic health record analysis demonstrate the effectiveness of our approach and highlight the limitations of existing L2D methods in multi-task scenarios. |
| title | A Two-Stage Learning-to-Defer Approach for Multi-Task Learning |
| topic | Machine Learning Human-Computer Interaction |
| url | https://arxiv.org/abs/2410.15729 |