Gareth Conduit has a track record of applying artificial intelligence to solve real-world problems.
The approach, originally developed for materials design at University of Cambridge, is now being commercialized by startup Intellegens in not only materials design, but also healthcare and drug discovery. Previously, Gareth had research interests in strongly correlated phenomena, in particular proposing a spin spiral state in the itinerant ferromagnet that was later observed in CeFePO.
Gareth's group is based at the University of Cambridge.
Green materials in less time: accelerated discovery with machine learning
We present a machine learning methodology, Alchemite, that exploits multiple sources of materials information: experimental data, physical laws, and computer simulations, to make the best possible predictions. We illustrate the approach with two case studies:
Firstly, we design and experimentally verify a nickel-base superalloy for direct laser deposition that could markedly improve the fuel efficiency and lower the carbon footprint of gas turbine engines. Starting from a training set comprising just eight results, the machine learning tool juxtaposes complementary material properties to circumvent missing data.
Secondly, we use Alchemite to guide the design of materials for batteries for use in electric vehicles. The proposed graphite-based electrodes are experimentally verified to have excellent cycle life and capacity in agreement with machine learning predictions.
Finally, we discuss further case studies that highlight the breadth of applications of the generic approach.
