Artificial Intelligence-Another modest cough from LSS

It’s not that we want to boast or strut about with an oh-so-clever-told-you-so lower sixth form grin on our faces. But we at LSS have been extolling the virtues of Artificial Intelligence in countless fields. (LSS 28 April 2020) And so we feel ever so slightly vindicated at today’s repro from Nature updates. The researchers in this article are using the power of AI to look for new materials to help reduce CO2 emissions. We pointed out how AI had helped to discover a new antimicrobial called halicin.* Either way, the implications for research into the Covid-19 pandemic are blatant, if our leaders could get their acts together. It’s amazing what people can achieve when they think, as opposed to screaming with their emotions.

here’s the Nature summary, which gives you the main paper:

Researchers on the hunt for new materials are increasingly turning to artificial intelligence (AI) and machine learning. Algorithms can predict the physical characteristics of selected crystal structures from first principles, and neural networks can use that information to make guesses about much larger gamuts of possible materials. In future, automated labs might help to make the materials more quickly. But even a robotic lab will need human overseers: synthesis still involves “a fair amount of artisanship,” says electrical engineer Ted Sargent. Wired | 6 min read
Source: Nature paper

Before you go off to read the paper, the team that did this looks sort of international and cooperative to us. Is that telling you something? Or maybe you’re the sort who thinks your football team should only use players from your own country. Well, it’s one way of doing things.

*J Stokes R Barzailly J Collins et al A deep learning approach to antibiotic discovery CELL vol 180 pp 688-702 Feb 20 2020

#artificialintelligence #nature #materialsscience #catalyst