4.7 Article

Deep Evolutionary Learning for Molecular Design

期刊

IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE
卷 17, 期 2, 页码 14-28

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/MCI.2022.3155308

关键词

Computational modeling; Social factors; Evolutionary computation; Statistics; Optimization; Deep learning; Generative adversarial networks; Molecular computing

资金

  1. Artificial Intelligence for Design Challenge Program at the National Research Council Canada
  2. Natural Sciences and Engineering Research Council of Canada (NSERC) [RGPIN-2021-03879]
  3. Vector Scholarship in AI - Vector Institute

向作者/读者索取更多资源

This paper proposes a prototypical deep evolutionary learning (DEL) process that integrates deep generative model and multi-objective evolutionary computation for molecular design. DEL enables evolutionary operations in the latent space of the generative model, generates promising novel molecular structures, and improves the generative model learning through fine-tuning. Experimental results show that DEL achieves improvement in property distributions and outperforms other baseline molecular optimization algorithms in generating samples. Additionally, comparisons with various deep generative models demonstrate the benefits of DEL in improving sample populations.
In this paper, a prototypical deep evolutionary learning (DEL) process is proposed to integrate deep generative model and multi-objective evolutionary computation for molecular design. Our approach enables (1) evolutionary operations in the latent space of the generative model, rather than the structural space, to generate promising novel molecular structures for the next evolutionary generation, and (2) generative model fine-tuning using newly generated high-quality samples. Thus, DEL implements a data-model co-evolution concept which improves both sample population and generative model learning. Experiments on public datasets indicate that the sample population obtained by DEL exhibits improvement on property distributions, and dominates samples generated by other baseline molecular optimization algorithms. Furthermore, comparisons with a range of deep generative models show that DEL is beneficial for improving sample populations.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据