Multi-Objective Latent Space Optimization of Generative Molecular Design Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911984240295936 |
|---|---|
| author | Abeer, A N M Nafiz Urban, Nathan Weil, M Ryan Alexander, Francis J. Yoon, Byung-Jun |
| author_facet | Abeer, A N M Nafiz Urban, Nathan Weil, M Ryan Alexander, Francis J. Yoon, Byung-Jun |
| contents | Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization. In this paper, we propose a multi-objective latent space optimization (LSO) method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2203_00526 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Multi-Objective Latent Space Optimization of Generative Molecular Design Models Abeer, A N M Nafiz Urban, Nathan Weil, M Ryan Alexander, Francis J. Yoon, Byung-Jun Machine Learning Biomolecules Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization. In this paper, we propose a multi-objective latent space optimization (LSO) method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties. |
| title | Multi-Objective Latent Space Optimization of Generative Molecular Design Models |
| topic | Machine Learning Biomolecules |
| url | https://arxiv.org/abs/2203.00526 |