Applied machine learning to accelerate development of formulations and chemical processes
Time & Location
Wednesday: 12.00 to 12.20, Stage 4
Dr Gareth ConduitCTO, Intellegens
About this presentation
We’ll discuss, with case study examples, the application of machine learning to accelerate discovery and development of formulations, chemicals, and chemical processes. Design and optimisation for these products and processes is time-consuming and expensive, with a high reliance on experimentation. Machine learning has potential for a dramatic speed-up, by narrowing-down the number of experiments required to find a solution to a much higher degree than conventional experimental design techniques. Machine learning also finds solutions that other approaches may miss. But there are some practical barriers. Machine learning often fails when faced with real-world, sparse, noisy experimental data. In this talk, we will introduce Alchemite, advanced deep learning software that overcomes this sparse data challenge, and review case studies of its use for applications including chemical processes, ink formulations, fragrances, drug discovery, plastics, and additive manufacturing.
Dr Gareth Conduit is Chief Technology Officer at Intellegens and a Royal Society Research Fellow at the University of Cambridge. He has a track record of applying machine learning to solve real-world problems, with research contacts held with companies spanning materials science to drug discovery. He is leading the development of the Alchemite software and of new machine learning methodologies for industrial R&D.