About

I am a Lecturer (Assistant Professor) in Machine Learning based at the School of Informatics within The University of Edinburgh. I also hold a Royal Academy of Engineering Research Fellowship on verifiable and robust meta-learning. I was previously a postdoc with Prof Timothy Hospedales, and a PhD student with Prof Bernhard Pfahringer.

I have a particular interest in developing and leveraging theoretical advances to improve how machine learning is done in practice. I am interested in applying machine learning to a variety of application domains, including signal processing, computer vision, natural language processing, robotics, and data mining. My current research is focused on developing machine learning systems that can seek out and/or leverage extra contextual information about the tasks to be solved, thus enabling one to learn more robust models with less data. Such contextual information could include examples of good models for related tasks, natural language descriptions of the phenomena being modelled, knowledge of underlying causal mechanisms, or other meta-data associated with tasks.

News

• 18/07/2023 Our paperon Quality Diversity ensembles via different invariances has been accepted at ICCV 2023!
• 01/04/2023 I have a fully funded PhD studentship available for both UK and international applicants! If you are interested in the research area outlined above, please get in contact with me ASAP.
• 27/02/2023 The Meta-Omnium benchmark for multi-task meta-learning has been accepted at CVPR 2023!
• 28/11/2022 I am co-organising the ICLR 2023 workshop What do we need for successful domain generalization?
• 30/09/2022 Why do Self-Supervised Models Transfer? Investigating the Impact of Invariance on Downstream Tasks has been accepted at BMVC 2022
• 01/09/2022 I am now a Royal Academy of Egineering Research Fellow!
• 20/07/2022 The NeurIPS 2021 competition track paper about the meta-learning competition is now published!
• 15/05/2022 Loss Function Learning for Domain Generalization by Implicit Gradient has been accepted at ICML 2022
• 06/05/2022 Our tutorial paper on self-supervised learning is now published in the IEEE Signal Processing Magazine!
• 15/04/2022 I will be co-organising a combined workshop and tutorial on meta-learning at ECML-PKDD 2022.
• 04/08/2022 The journal version of our paper analysing cross-domain few-shot learning is now published in the Journal of the Royal Society of New Zealand.
• 24/03/2022 Our work on meta-learning mirror descent optimisers has been accepted at the ICLR 2023 GroundedML workshop—extended version coming soon!
• 22/07/2021 Our papers on MetaQDA and meta-learning loss functions for noisy label problems have been accepted at ICCV 2021!

Joining my Group

If you are interested joining my research group as an undergraduate already enrolled at University of Edinburgh, or as a postgraduate student, please feel free to get in contact with me via email. You will need to provide me with an indication of a specific topic you are interested in working on.