About

I am a final year PhD student in the Department of Chemistry at University College London (UCL), using data-driven methods to find novel polar photovoltaics.

I completed my undergraduate integrated Master’s degree in Natural Sciences at the University of Cambridge, with a research project under Prof. Clare Grey and Dr Sunita Dey synthesising and characterising borate-based cathodes for rechargeable magnesium ion batteries (RMBs).
Next, I studied for an MSc in Materials and Molecular Modelling at UCL to refine my computational skills, including a research project with Prof. David Scanlon in which I used plane-wave DFT (in VASP) to study the thermoelectric properties of Ba2AgIO6.
During my PhD, my focus has shifted to using machine learning methods for materials discovery, with a particular focus on polar photovoltaics. Recent projects have included comparing the carbon cost of various ML/DFT workflows for PV discovery, using Fused Gromov-Wasserstein (FGW) distances for similarity learning, and compressing absorption spectra using disentangling autoencoders (DAEs).
Outside of chemistry and machine learning, I like running, lifting heavy things, cooking, playing chess, reading, and learning languages.
See my publications and posts for more details about my work, my CV for my full education and employment history, or contact me at matthewahwalker@gmail.com for collaborations.