Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909121482063872 |
|---|---|
| author | A V, Arun Kumar Shilton, Alistair Gupta, Sunil Rana, Santu Greenhill, Stewart Venkatesh, Svetha |
| author_facet | A V, Arun Kumar Shilton, Alistair Gupta, Sunil Rana, Santu Greenhill, Stewart Venkatesh, Svetha |
| contents | Experimental (design) optimization is a key driver in designing and discovering new products and processes. Bayesian Optimization (BO) is an effective tool for optimizing expensive and black-box experimental design processes. While Bayesian optimization is a principled data-driven approach to experimental optimization, it learns everything from scratch and could greatly benefit from the expertise of its human (domain) experts who often reason about systems at different abstraction levels using physical properties that are not necessarily directly measured (or measurable). In this paper, we propose a human-AI collaborative Bayesian framework to incorporate expert preferences about unmeasured abstract properties into the surrogate modeling to further boost the performance of BO. We provide an efficient strategy that can also handle any incorrect/misleading expert bias in preferential judgments. We discuss the convergence behavior of our proposed framework. Our experimental results involving synthetic functions and real-world datasets show the superiority of our method against the baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_17343 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties A V, Arun Kumar Shilton, Alistair Gupta, Sunil Rana, Santu Greenhill, Stewart Venkatesh, Svetha Machine Learning Experimental (design) optimization is a key driver in designing and discovering new products and processes. Bayesian Optimization (BO) is an effective tool for optimizing expensive and black-box experimental design processes. While Bayesian optimization is a principled data-driven approach to experimental optimization, it learns everything from scratch and could greatly benefit from the expertise of its human (domain) experts who often reason about systems at different abstraction levels using physical properties that are not necessarily directly measured (or measurable). In this paper, we propose a human-AI collaborative Bayesian framework to incorporate expert preferences about unmeasured abstract properties into the surrogate modeling to further boost the performance of BO. We provide an efficient strategy that can also handle any incorrect/misleading expert bias in preferential judgments. We discuss the convergence behavior of our proposed framework. Our experimental results involving synthetic functions and real-world datasets show the superiority of our method against the baselines. |
| title | Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.17343 |