My mother always tells the story of how I learnt to type my name on a computer before I could put pen to paper. I grew up with a love of computers and am not ashamed to say that the topic of artificial intelligence (AI) – covering the gamut of machine learning, and deep learning – is a particular passion. You can imagine my delight, therefore, when I came across not one but two recent articles on how the future of materials science and AI may be intertwined.
There are real-world application examples aplenty in a post on SingularityHub, provocatively titled ‘The Wild New Materials of the Future Will Be Discovered With AI’. According to the article, a team at Stanford University has been using machine learning to develop better electrolytes for lithium-ion batteries. Evan Reed, assistant professor of Materials Science and Engineering commented that the team has developed a machine learning model that has been outperforming experts’ intuition when predicting which materials to use. Meanwhile at the University of Maryland, a research scientist named Valentin Stanev has applied machine learning to the question of whether a material has the potential to be a superconductor.
Taking things from another angle, is the Advanced Science News article: ‘Artificial Intelligence Models New Functional Materials’. The article highlights that a new computational model (powered by AI) is attempting to predict the functional properties of new materials, with the aim of saving on raw materials, and effort in the lab. ‘A research team working in China and the UK has combined artificial neural networks and evolutionary computation to generate the model, and then applied it to determine the parameters required to form self-cleaning, superhydrophobic surfaces’, the article states.
If we’re on the cusp of big data within materials science, it’s safe to say that AI will push us over that threshold – and the data generation possibilities are impressive. Of course, that will inevitably lead to the questions of how best to collate, analyse, and apply that explosion of data, whilst maintaining its pedigree. Here’s where materials information management technologies like GRANTA MI can help. Find out how: https://www.grantadesign.com/products/mi/index.htm
Latest posts by Beth Harlen (see all)
- From fashion to flight: the diverse users of CES Selector - 24th May 2018
- Is the future of materials science and AI intertwined? - 16th April 2018
- Composite complications – why hybrid materials are a particular materials information menace - 5th April 2018