Algorithms with Calibrated Machine Learning Predictions
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917361217437696 |
|---|---|
| author | Shen, Judy Hanwen Vitercik, Ellen Wikum, Anders |
| author_facet | Shen, Judy Hanwen Vitercik, Ellen Wikum, Anders |
| contents | The field of algorithms with predictions incorporates machine learning advice in the design of online algorithms to improve real-world performance. A central consideration is the extent to which predictions can be trusted -- while existing approaches often require users to specify an aggregate trust level, modern machine learning models can provide estimates of prediction-level uncertainty. In this paper, we propose calibration as a principled and practical tool to bridge this gap, demonstrating the benefits of calibrated advice through two case studies: the ski rental and online job scheduling problems. For ski rental, we design an algorithm that achieves near-optimal prediction-dependent performance and prove that, in high-variance settings, calibrated advice offers more effective guidance than alternative methods for uncertainty quantification. For job scheduling, we demonstrate that using a calibrated predictor leads to significant performance improvements over existing methods. Evaluations on real-world data validate our theoretical findings, highlighting the practical impact of calibration for algorithms with predictions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_02861 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Algorithms with Calibrated Machine Learning Predictions Shen, Judy Hanwen Vitercik, Ellen Wikum, Anders Machine Learning Data Structures and Algorithms The field of algorithms with predictions incorporates machine learning advice in the design of online algorithms to improve real-world performance. A central consideration is the extent to which predictions can be trusted -- while existing approaches often require users to specify an aggregate trust level, modern machine learning models can provide estimates of prediction-level uncertainty. In this paper, we propose calibration as a principled and practical tool to bridge this gap, demonstrating the benefits of calibrated advice through two case studies: the ski rental and online job scheduling problems. For ski rental, we design an algorithm that achieves near-optimal prediction-dependent performance and prove that, in high-variance settings, calibrated advice offers more effective guidance than alternative methods for uncertainty quantification. For job scheduling, we demonstrate that using a calibrated predictor leads to significant performance improvements over existing methods. Evaluations on real-world data validate our theoretical findings, highlighting the practical impact of calibration for algorithms with predictions. |
| title | Algorithms with Calibrated Machine Learning Predictions |
| topic | Machine Learning Data Structures and Algorithms |
| url | https://arxiv.org/abs/2502.02861 |