Towards Geometric Deep Learning IV: Chemical Precursors of GNNs

Ifthe <a href="https://towardsdatascience.com/towards-geometric-deep-learning-i-on-the-shoulders-of-giants-726c205860f5?sk=fd04bfaab732177ba7b4d7da90d88e9e" rel="noopener" target="_blank">history of symmetry</a> is tightly intertwined with physics, the history of graph neural networks, a “poster child” of Geometric Deep Learning, has roots in another branch of natural science: chemistry. Chemistry has historically been — and still is — one of the most data-intensive academic disciplines. The emergence of modern chemistry in the eighteenth century resulted in the rapid growth of known chemical compounds and an early need for their organisation. This role was initially played by periodicals such as the Chemisches Zentralblatt [1] and “chemical dictionaries” like the Gmelins Handbuch der anorganischen Chemie (an early compendium of inorganic compounds <a href="https://www.chemistryworld.com/features/200-years-of-gmelins-handbook/3007265.article" rel="noopener ugc nofollow" target="_blank">first published in 1817</a> [2]) and Beilsteins Handbuch der organischen Chemie (a similar effort for organic chemistry) — all initially published in German, which was the dominant language of science until the early 20th century. In the English-speaking world, the Chemical Abstracts Service (CAS) was created in 1907 and has gradually become the central repository for the world’s published chemical information… <a href="https://towardsdatascience.com/towards-geometric-deep-learning-iv-chemical-precursors-of-gnns-11273d74125">Read More</a>