Artificial Intelligence at the Interface of Chemistry and Mathematics

仅仅通过计算机模拟，能够预先选择值得进行实验测试的药物，预测特定化合物对DNA的损伤机理，或者设计具有特定性质的材料，无论是在技术进步、健康还是经济和节能方面，都是全社会不容置疑的兴趣。尽管化学和生物分子科学取得了所有重要进展，但仍然存在一些基本问题，阻碍了此类计算的结果和结论的完全可靠性，无论它们是通过量子计算还是机器学习算法获得的。最先进的机器学习方法和量子化学计算提供了令人印象深刻的结果，为科学、技术和社会开辟新的视角。然而，它们依赖于大量的计算资源来处理巨大的手动注释数据集。相应的能源消耗和人力成本是不可持续的。无论是用生物、化学、物理还是工程学的语言来表述，所有这些悬而未决的问题的核心挑战都是数学。可持续和高效的计算算法的发展需要超越边界，并结合来自不同科学领域的复杂知识。Gerolin提出的CRC研究计划开发了新的数学和计算工具，将数学、机器学习、计算和理论化学的概念结合在一起，目标是打破这些理论和计算的限制。该计划的成果可应用于过渡金属(催化的主力)、分散相互作用(例如DNA中的氢键相互作用)、化学和药物发现。开发的严谨研究可以帮助其他研究同一问题的研究人员消除一些猜测和经验主义，从而产生更可持续的科学。这种跨学科的方法极有希望增进科学界之间的相互理解。

Being able, by sole means of computer simulations, to make a preselection of pharmaceutical drugs that deserve to be tested experimentally, to predict the mechanisms of DNA damage by specific compounds, or to design materials with specific properties, is of unquestionable interest for the whole society, in terms of both technological progress, health and economic and energetic saving.Despite all important advances in chemical and biomolecular sciences, there are still basic unsolved problems that hamper a complete reliability of the results and conclusions of such calculations, whether they are obtained via quantum computations or Machine Learning algorithms.The state-of-the-art Machine Learning methods and Quantum Chemistry calculations provide impressive results, opening new perspectives for science, technology, and society. However, they rely on massive computational resources to process huge manually annotated data-sets. The corresponding costs in terms of energy consumption and human efforts are not sustainable.Whether phrased in the language of biology, chemistry, physics, or engineering, the central challenge to all these unsolved problems is mathematical. The development of sustainable and efficient computational algorithms require transcending boundaries and to combine sophisticate knowledge from different scientific areas.Dr. Gerolin's proposed CRC research program develop novel mathematical and computational tools combining concepts from Mathematics, Machine Learning, Computational and Theoretical Chemistry with the goal to break such theoretical and computational limitations. The outcomes of this program can be applied to transition metals (which are the workhorse of catalysis), dispersion interactions (e.g. hydrogen-bonding interaction in DNA), chemical and drug discovery. The rigorous studies developed could help other researchers who work on the same problem to eliminate some guessing and empiricism, leading to a more sustainable science. The interdisciplinary approach holds tremendous promise to increase the mutual understanding between scientific communities.