另一方面，盡管大多數(shù)定性（或分類）變量（如化學(xué)元素、化學(xué)成分）比定量變量更容易獲得，但在自動(dòng)材料設(shè)計(jì)中直接將定性變量作為設(shè)計(jì)變量的一部分是一個(gè)挑戰(zhàn)。

金屬有機(jī)框架（MOFs）就是這類材料系統(tǒng)的一個(gè)例子。MOFs是一類多孔結(jié)晶材料，廣泛用于氣體儲(chǔ)存、氣體分離和催化。由于其高度可調(diào)性，MOFs被視為解決不同應(yīng)用問(wèn)題的潛在方案，例如二氧化碳（CO₂）的捕集和分離。然而，由于MOF構(gòu)建塊及其組合方式的多樣性，候選材料數(shù)量級(jí)過(guò)高。

因此，實(shí)驗(yàn)所需的時(shí)間和資源太高，人們已經(jīng)開(kāi)始使用機(jī)器學(xué)習(xí)來(lái)加速材料系統(tǒng)的設(shè)計(jì)和開(kāi)發(fā)。但現(xiàn)有的方法通常依賴于大量的數(shù)據(jù)集和高維物理描述符來(lái)表示材料設(shè)計(jì)空間。這些機(jī)器學(xué)習(xí)模型既耗時(shí)，泛化性又不強(qiáng)，通常不能遷移到不同的設(shè)計(jì)目標(biāo)上。

Fig. 4 The LVGP-BO results for the Reduced Design Space (RDS) exploration.

來(lái)自美國(guó)西北大學(xué)機(jī)械系的Yigitcan Comlek等，提出了一套潛在變量高斯過(guò)程多目標(biāo)批量貝葉斯優(yōu)化（LVGP-MOBBO）框架，以直接從構(gòu)建材料的構(gòu)建塊中快速設(shè)計(jì)優(yōu)越的MOFs。

他們使用了已有的定性MOFs建筑塊信息，構(gòu)建了一個(gè)可解釋的LVGP模型，在MOBBO的輔助下，自適應(yīng)地引導(dǎo)CO₂捕獲和分離性能較好的MOFs。

他們通過(guò)整合批量貝葉斯優(yōu)化，無(wú)描述符的LVGP也可以有效地?cái)U(kuò)展到具有大量級(jí)別的應(yīng)用。通過(guò)LVGP預(yù)測(cè)具有看不見(jiàn)構(gòu)建塊的MOFs的特性是一個(gè)很有前途的研究領(lǐng)域。

該框架的一個(gè)有趣的應(yīng)用是將涉及到通過(guò)自主實(shí)驗(yàn)研究進(jìn)行材料設(shè)計(jì)和開(kāi)發(fā)。由于在LVGP-MOBBO中沒(méi)有人為干預(yù)，而且實(shí)驗(yàn)輸入可以是定性和定量的，在這里提出的方法可以幫助研究人員有效地指導(dǎo)實(shí)驗(yàn)。

Editorial Summary

With recent advances in machine learning (ML), material system design and development has undergone rapid acceleration. However, one of the major challenges in applying ML to material system design lies in finding the appropriate design representations. Most material design applications take advantage of quantitative (or numerical) design variables to represent material systems. In most cases, these quantitative descriptors (features) require either expert knowledge or data analysis to find the most appropriate ones. On the other hand, although most qualitative (or categorical) variables (e.g., chemical elements, chemical compositions) are more accessible than quantitative variables, it is challenging to directly include qualitative variables as a part of the design variables in automated materials design. Metal-organic frameworks (MOFs) are an example of such materials systems.

MOFs are a class of porous crystalline materials that have been used extensively for gas storage, gas separation, and catalysis. Because of their highly tunable nature, MOFs have been looked at as a potential solution for different applications such as CO₂ capture and separation. However, the versatility and different possible combinations of the MOF building blocks lead to millions of candidates. Due to the high experimental cost, both in time and resources, machine learning has been used to accelerate material system design and development. However, the existing approaches usually rely on large data sets and high-dimensional physical descriptors to represent the material design space. These processes can be both time consuming and property specific, meaning that the ML models and descriptors are often not transferable to different design objectives.?

Yigitcan Comlek et al. from the Department of Mechanical Engineering, Northwestern University, presented a Latent Variable Gaussian Process Multi-Objective Batch Bayesian Optimization (LVGP-MOBBO) framework to perform rapid design of superior MOFs directly from the building blocks that construct the material. They took advantage of the readily available qualitative building block information that is used to construct the MOFs and built an interpretable LVGP surrogate model that cooperates with MOBBO to adaptively lead towards promising MOF candidates for CO₂ capture and separation. With the integration of batch BO, descriptor-free LVGP can be effectively extended to applications with substantial number of levels. To predict the properties of MOFs with unseen building blocks through LVGP is a promising area of research. The interesting application of this framework would involve performing materials design and development through autonomous experimentation studies. As there is no human intervention in LVGP-MOBBO, and the experimental inputs can be both qualitative and quantitative, the method presented in this work can help researchers guide their experiments efficiently.

原文Abstract及其翻譯

Rapid design of top-performing metal-organic frameworks with qualitative representations of building blocks （快速設(shè)計(jì)具有定性表示構(gòu)建塊的性能最佳的金屬有機(jī)框架）

Yigitcan Comlek, Thang Duc Pham, Randall Q. Snurr & Wei Chen

Abstract

Data-driven materials design often encounters challenges where systems possess qualitative (categorical) information. Specifically, representing Metal-organic frameworks (MOFs) through different building blocks poses a challenge for designers to incorporate qualitative information into design optimization, and leads to a combinatorial challenge, with large number of MOFs that could be explored. In this work, we integrated Latent Variable Gaussian Process (LVGP) and Multi-Objective Batch-Bayesian Optimization (MOBBO) to identify top-performing MOFs adaptively, autonomously, and efficiently. We showcased that our method (i) requires no specific physical descriptors and only uses building blocks that construct the MOFs for global optimization through qualitative representations, (ii) is application and property independent, and (iii) provides an interpretable model of building blocks with physical justification. By searching only ~1% of the design space, LVGP-MOBBO identified all MOFs on the Pareto front and 97% of the 50 top-performing designs for the CO₂?working capacity and CO₂/N₂?selectivity properties.

摘要?

定性(分類)信息的系統(tǒng)通常會(huì)給數(shù)據(jù)驅(qū)動(dòng)材料設(shè)計(jì)帶來(lái)挑戰(zhàn)。特別地，通過(guò)不同的構(gòu)建塊來(lái)表示金屬有機(jī)框架（MOFs）給設(shè)計(jì)者將定性信息納入設(shè)計(jì)優(yōu)化帶來(lái)了挑戰(zhàn)，同時(shí)也帶來(lái)了一個(gè)組合型的挑戰(zhàn)，即設(shè)計(jì)者們能夠探索的MOFs太多。在本工作中，我們集成了隱變量高斯過(guò)程（LVGP）和多目標(biāo)批量-貝葉斯優(yōu)化（MOBBO），以自適應(yīng)、自主和高效地識(shí)別性能最好的MOFs。我們展示了我們的方法（i）不需要特定的物理描述符，只使用構(gòu)建塊來(lái)構(gòu)建MOFs，通過(guò)定性表示進(jìn)行全局優(yōu)化，（ii）應(yīng)用和屬性獨(dú)立，（iii）提供了一個(gè)具有物理證明的可解釋構(gòu)建塊模型。通過(guò)僅搜索約1%的設(shè)計(jì)空間，LVGP-MOBBO識(shí)別了Pareto前沿的所有MOFs，在目前50種CO₂吸收效率與CO₂/N₂選擇性能最好的設(shè)計(jì)中搜索出了97%的樣本。