My DPhil is about the development of formal methods for multi-agent and machine learning systems in order to help build provably safe and beneficial AI. I am supervised by Michael Wooldridge, Julian Gutierrez, and Alessandro Abate, and am also a DPhil Affiliate at the Future of Humanity Institute. Before coming to Oxford I worked as an intern on Imandra and was a research assistant for Jacques Fleuriot at the University of Edinburgh, where I completed my MSc in artificial intelligence under the supervision of Vaishak Belle. Prior to this I studied for my BSc in mathematics and philosophy at the University of Warwick with Walter Dean. For information about my previous work see some of the links below, or alternatively my CV.
My research interests are broad but mostly centre around the combination of statistical and symbolic AI, and in particular how this can be used to improve robustness, explainability, and safety. Particular topics I have worked on before include statistical relational learning, formal verification of learnt models, and deep symbolic reinforcement learning. Much of my work is also concerned with the problem of how agents represent and reason about preferences in the face of uncertainty. Examples of this include an ongoing project on learning models of cognitive biases using inverse reinforcement learning and my master's dissertation on computational frameworks for moral decision-making. Thirdly and finally, I also take an interest in the governance, ethics, and societal impact of AI, though these are not topics I am actively researching at present.
Outside of academia my main passion is music, but I also love film and art. I like to travel whenever I get the opportunity, especially to Scandinavia (where I previously lived), and in my spare time I enjoy reading, cooking, drawing, and clubbing. Ethically and politically I consider myself an effective altruist, a humanist, and a fabian (more generally, a democratic socialist).
15.02.20 - I'll be giving a short presentation on modelling agent incentives, and joining a panel discussion on AI alignment (alongside Rohin Shah and Michael Cohen) at next week's AI Ethics Meetup in London.
09.12.19 - My previous supervisor and co-author Vaishak Belle will be presenting our poster on Tractable Probabilistic Models for Moral Responsibility at the Knowledge Representation & Reasoning Meets Machine Learning (KR2ML) Workshop at NeurIPS on the 13th of December.
13.09.19 - This weekend I'll be in Switzerland attending the AI Governance Careers workshop at ETH Zurich.
01.07.19 - I'm very excited to be joining Aesthetic Integration as a research and engineering intern for the coming months in Edinburgh, where I'll be working on Imandra - a cloud-native automated reasoning engine.
17.05.19 - I have been invited to give a short presentation and join a panel discussion on 'Fair machines: Student perspective on Data Justice and Ethics' as part of the Data Justice Week taking place in Edinburgh between the 20th-24th of May.
26.04.19 - I will be attending the third AI Safety Camp in Ávila, Spain for the next week and a half, where I am working on a group project to improve preference elicitation by first learning models of biased and mistaken behaviour in agents.
09.04.19 - A short version of my first paper, Deep Tractable Probabilistic Models for Moral Responsibility, has been accepted for presentation at the Human-Like Computing Third Wave of AI Workshop (3AI-HLC 2019) taking place at Imperial College London on the 26th of April.
03.03.19 - I am delighted to have accepted an offer to study for a DPhil in computer science at the University of Oxford later this year, which will be generously funded by the EPSRC Doctoral Training Partnership.