Bio
Laura is a research associate in cryptographic machine learning in the Defence and Security programme. She started out in laboratory neuroscience combining electrophysiology with optogenetics to explain the coordination of place cell ensembles at the Neural Circuits and Memory Lab, before developing value functions to depart from expected utility theory for human-aligned reinforcement learning at Mila, the Quebec Artificial Intelligence Institute. Her work at the Turing focuses on accelerating the development of scalable privacy-enhancing technologies by adapting tensor networks as an alternative to neural networks for use on homomorphically encrypted data. She has also written about maths and science for a general audience in MIT Technology Review and Nautilus and was a journalist-in-residence at the Kavli Institute for Theoretical Physics.