Gareth Conduit

Gareth Conduit

Royal Society University Research Fellow in the Theory of Condensed Matter Group, at the University of Cambridge. The group develops and applies machine learning to design new materials & drugs and researches quantum phenomena. The machine learning technology Alchemite™ is commercialized for materials and industrial chemicals design through Intellegens; as the drug design suite Cerella™ by Optibrium; and for additive manufacturing through ANSYS Granta. Further details can be found in the research stories below, news coverage, recent talks, and publications.

Material design

Alloy microstructure
Microstructure of nickel-base superalloy for additive manufacturing designed with Alchemite™. Materials & Design 168, 107644 (2019).

Through the stone, bronze, and iron ages materials have chronicled human history. Despite the central importance of materials in enabling technologies, even today materials are discovered through experimental driven trial and improvement. We proposed, implemented, and employed a machine learning tool, Alchemite™, that has specialist capabilities to handle sparse experimental data.

Alchemite™ was first used with Rolls Royce to design three alloy families [1, 2, 3], whose physical properties have been experimentally verified to match or exceed commercially available alternatives. Alchemite™ can also juxtapose experimental data with first principles computer simulations, demonstrated to explore new lubricants, metal organic frameworks, and battery lifetime predictions. The tool is commercialized by Intellegens as the product Alchemite™ Analytics, and is also available for additive manufacturing through partners ANSYS Granta.

Drug design

Open Source Malaria competition molecule
Molecule designed with Alchemite™ that won the Open Source Malaria competition. Journal of Medicinal Chemistry 64, 16450 (2021).

Designing drugs is a huge computational challenge: finding the drug molecule that will correctly affect the behavior of all 10,000 proteins in the human body. The search is complicated by the dearth of information: just 0.05% of the known drug molecule-protein activity levels are actually known. The machine learning tool, Alchemite™, developed for materials design is perfectly suited for handling sparse data. Identification and exploitation of the correlations between activity levels of different proteins allows Alchemite™ to deliver accurate predictions for drug efficacy.

The additional insights offered by Alchemite™ reduces the need to perform additional experiments, cutting the cost and accelerating the discovery of drugs. Alchemite™ achieves the best prediction accuracy ever seen on a benchmarking data set published by Novartis, and won the Open Source Malaria competition with the only molecule experimentally verified to be active. Alchemite™ machine learning technology is marketed to industry as the Cerella™ suite by Optibrium and is used by multinational pharmaceutical companies to accelerate their drug discovery programs. Alchemite™ is also used in several commercial and academic projects.

Magnetic spin spiral

Magnetic spin spiral
Three-step procedure for the formation of a magnetic spin spiral. Phys. Rev. A 82, 043604 (2010).

All of the electrons in a materials solid interact with each other, so that when one electron moves it pushes all of the other electrons in the material. Furthermore, electrons are quantum particles so they obey the counter-intuitive laws of quantum mechanics. The juxtaposition of many-body interactions and quantum mechanics leads to exotic phenomena including ferromagnetism and quantum mechanics.

An electron gas with contact repulsive interactions is a deceptively simple interacting system, yet it displays a remarkably rich range of phenomena. At mean-field level the electron gas was predicted by Stoner to undergo a mean-field transition into an itinerant ferromagnet. This phase has never been cleanly observed in the solid state, however our study underpinned the first experimental exploration in an ultracold atomic gas.

Moreover, quantum fluctuations mean that a variety of other inhomogeneous magnetic states are energetically favorable, with our suggestion of a spin spiral state being first observed in CeFePO in 2012.

Few trapped atoms

Atomic orbitals
Orbitals of two up-spin and one down-spin fermion with repulsive interactions. Phys. Rev. Lett. 111, 045301 (2013).

A few-fermion ultracold atom system presents an alternative arena to study strongly interacting fermions. Ultracold atoms offer fully tunable interactions, and being few fermions permits exact analytical or computational solutions.

Our three-part study covered the most important phases of matter. Repulsive interactions provides an analogy of itinerant ferromagnetism, attractive interactions explores superfluidity, and exchange interactions the crossover from a conductor to an insulator.

Our study was followed by the experimental realization and characterization of a few-atom Fermi sea by the Jochim group. The study showed that the crossover to many-body physics takes place with just six fermions, in agreement with theoretical predictions, and furthermore delivered new insights into the physics of repulsive interactions. The work laid the foundation for few trapped atoms to be an ideal playground to develop intuition and understanding of many-body physics.