Synergy Between Expert and Machine-Learning Approaches Allows for Improved Retrosynthetic Planning
Synergy Between Expert and Machine-Learning Approaches Allows for Improved Retrosynthetic Planning
复制标题
DOI：
10.1002/anie.201912083
复制
发表时间：
2019-11-21
影响因子：
16.6
通讯作者：
Grzybowski, Bartosz A.
中科院分区：
文献类型：
--
作者：
Badowski, Tomasz;Gajewska, Ewa P.;Grzybowski, Bartosz A.
当计算机计划多步骤合成时，他们可以依靠专家知识，也可以依靠机器从大型反应储存库中提取的信息。这两种方法都存在评估反应选择的函数不完善的问题：专家函数是基于化学直觉的启发式方法，而机器学习(ML)依赖于神经网络(NNS)，神经网络只能对流行的反应类型做出有意义的预测。本文表明，专家方法和最大似然方法可以是协同的-具体地说，当神经网络基于与高质量的专家编码的反应规则匹配的文献数据进行训练时，它们比单独使用这两种方法获得更高的合成精度，更重要的是，它们还可以处理罕见/专门的反应类型。
When computers plan multistep syntheses, they can rely either on expert knowledge or information machine-extracted from large reaction repositories. Both approaches suffer from imperfect functions evaluating reaction choices: expert functions are heuristics based on chemical intuition, whereas machine learning (ML) relies on neural networks (NNs) that can make meaningful predictions only about popular reaction types. This paper shows that expert and ML approaches can be synergistic-specifically, when NNs are trained on literature data matched onto high-quality, expert-coded reaction rules, they achieve higher synthetic accuracy than either of the methods alone and, importantly, can also handle rare/specialized reaction types.