Generative Adversarial Model-Based Optimization via Source Critic Regularization
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914957203865600 |
|---|---|
| author | Yao, Michael S. Zeng, Yimeng Bastani, Hamsa Gardner, Jacob Gee, James C. Bastani, Osbert |
| author_facet | Yao, Michael S. Zeng, Yimeng Bastani, Hamsa Gardner, Jacob Gee, James C. Bastani, Osbert |
| contents | Offline model-based optimization seeks to optimize against a learned surrogate model without querying the true oracle objective function during optimization. Such tasks are commonly encountered in protein design, robotics, and clinical medicine where evaluating the oracle function is prohibitively expensive. However, inaccurate surrogate model predictions are frequently encountered along offline optimization trajectories. To address this limitation, we propose generative adversarial model-based optimization using adaptive source critic regularization (aSCR) -- a task- and optimizer- agnostic framework for constraining the optimization trajectory to regions of the design space where the surrogate function is reliable. We propose a computationally tractable algorithm to dynamically adjust the strength of this constraint, and show how leveraging aSCR with standard Bayesian optimization outperforms existing methods on a suite of offline generative design tasks. Our code is available at https://github.com/michael-s-yao/gabo |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_06532 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generative Adversarial Model-Based Optimization via Source Critic Regularization Yao, Michael S. Zeng, Yimeng Bastani, Hamsa Gardner, Jacob Gee, James C. Bastani, Osbert Machine Learning Artificial Intelligence Offline model-based optimization seeks to optimize against a learned surrogate model without querying the true oracle objective function during optimization. Such tasks are commonly encountered in protein design, robotics, and clinical medicine where evaluating the oracle function is prohibitively expensive. However, inaccurate surrogate model predictions are frequently encountered along offline optimization trajectories. To address this limitation, we propose generative adversarial model-based optimization using adaptive source critic regularization (aSCR) -- a task- and optimizer- agnostic framework for constraining the optimization trajectory to regions of the design space where the surrogate function is reliable. We propose a computationally tractable algorithm to dynamically adjust the strength of this constraint, and show how leveraging aSCR with standard Bayesian optimization outperforms existing methods on a suite of offline generative design tasks. Our code is available at https://github.com/michael-s-yao/gabo |
| title | Generative Adversarial Model-Based Optimization via Source Critic Regularization |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2402.06532 |