I am a fourth year PhD student at UCL in the Centre for Artificial Intelligence, supervised by François-Xavier Briol and Marc Deisenroth.

I'm currently interested in making large machine learning models more efficient, via both modelling and systems approaches, and developing better software to support researchers and practitioners. I am also excited about working on tools which are useful for climate change mitigation and adaptation.

Feel free to get in touch via email!

Bio

2020 -

PhD candidate in Computer Science UCL Supervised by François-Xavier Briol and Marc Deisenroth.

2019 - 2020

Research Assistant in Bayesian Deep Learning University of Oxford Oxford Applied and Theoretical Machine Learning (OATML) group, supervised by Yarin Gal

2018 - 2019

MSc in Computer Science University of Oxford Dissertation on "A comparison of uncertainty estimates for safe exploration in machine learning", supervised by Zac Kenton, Tim G. J. Rudner, Yarin Gal

2016 - 2018

Software Engineer Google Encryption of Android backup data using a client-side key, released in Android 9

2013 - 2016

BA in Computer Science University of Cambridge

Research and code

• No Train No Gain: Revisiting Efficient Training Algorithms For Transformer-based Language Models Jean Kaddour*, Oscar Key*, Piotr Nawrot, Pasquale Minervini, Matt J. Kusner arXiv code NeurIPS 2023 poster
• Interlocking Backpropagation: Improving depthwise model-parallelism Aidan N. Gomez*, Oscar Key*, Kuba Perlin, Stephen Gou, Nick Frosst, Jeff Dean, Yarin Gal arXiv code JMLR
• Optimally-Weighted Estimators of the Maximum Mean Discrepancy for Likelihood-Free Inference Ayush Bharti, Masha Naslidnyk, Oscar Key, Samuel Kaski, François-Xavier Briol arXiv code ICML 2023 poster
• Towards Healing the Blindness of Score Matching Mingtian Zhang, Oscar Key, Peter Hayes, David Barber, Brooks Paige, François-Xavier Briol arXiv Workshop on Score-Based Methods NeurIPS 2022
• Composite Goodness-of-fit Tests with Kernels Oscar Key*, Arthur Gretton, François-Xavier Briol, Tamara Fernandez* arXiv code talk Your Model is Wrong Workshop NeurIPS 2021
• On Feature Collapse and Deep Kernel Learning for Single Forward Pass Joost van Amersfoort, Lewis Smith, Andrew Jesson, Oscar Key, Yarin Gal arXiv code Bayesian Deep Learning Workshop at NeurIPS 2021
• On Signal-to-Noise Ratio Issues in Variational Inference for Deep Gaussian Processes Tim G. J. Rudner*, Oscar Key*, Yarin Gal, Tom Rainforth arXiv code ICML 2021 poster
• Generating Interpretable Counterfactual Explanations By Implicit Minimisation of Epistemic and Aleatoric Uncertainties Lisa Schut*, Oscar Key*, Rory McGrath, Luca Costabello, Bogdan Sacaleanu, Medb Corcoran, Yarin Gal arXiv code AISTATS 2021 poster ML4H Workshop at NeurIPS 2020
• PyTorch library for constrained MPC, using the cross-entropy method code