Publications

We do research in both theoretical machine learning, as well as applications of machine learning in various domains, including medical, astronomy, autonomous driving, and more.

Recent publications:

	07/2024: "AI models collapse when trained on recursively generated data" (Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao, Nicolas Papernot, Ross Anderson, Yarin Gal).
	07/2024: "Variational Inference Failures Under Model Symmetries: Permutation Invariant Posteriors for Bayesian Neural Networks" (Yoav Gelberg, Tycho F.A. van der Ouderaa, Mark van der Wilk, Yarin Gal).
	06/2024: "Detecting hallucinations in large language models using semantic entropy" (Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, Yarin Gal).

Group Blog

Recent group blog posts:

	19 Jun 2024: Detecting hallucinations in large language models using semantic entropy (Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, Yarin Gal)
	25 Nov 2022: OATML Conference papers at NeurIPS 2022 (Yarin Gal, Freddie Kalaitzis, Shreshth Malik, Lorenz Kuhn, Gunshi Gupta, Jannik Kossen, Pascal Notin, Andrew Jesson, Panagiotis Tigas, Tim G. J. Rudner, Sebastian Farquhar, Ilia Shumailov)
	17 Jul 2022: OATML at ICML 2022 (Sören Mindermann, Jan Brauner, Muhammed Razzak, Andreas Kirsch, Aidan Gomez, Sebastian Farquhar, Pascal Notin, Tim G. J. Rudner, Freddie Bickford Smith, Neil Band, Panagiotis Tigas, Andrew Jesson, Lars Holdijk, Joost van Amersfoort, Kelsey Doerksen, Jannik Kossen, Yarin Gal)
	22 Mar 2022: Itô and Stratonovich; a guide for the perplexed (Lewis Smith)

Reproducibility and Code

One of our lab's missions is to contribute to the reproducibility effort. Here we provide code for our own research, as well as reproductions of works by others (e.g. ones we implemented as baselines as part of our own research).

Recent code release:

	07/2024: AI models collapse when trained on recursively generated data (Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao, Nicolas Papernot, Ross Anderson, Yarin Gal).
	06/2024: Detecting hallucinations in large language models using semantic entropy (Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, Yarin Gal).
	04/2023: Bayesian active learning with EPIG data acquisition (Freddie Bickford Smith).
	05/2022: Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval (Pascal Notin).
	12/2021: Improving black-box optimization in VAE latent space using decoder uncertainty (Pascal Notin).

Group Members

We are researchers coming from varied backgrounds, including Computer Science, Maths & Stats, Engineering, and Physics.

We come from academia (Oxford, Cambridge, MILA, Manchester, U of Amsterdam, U of Toronto, U of Cape Town, Yale, and others) and industry (Google, DeepMind, Twitter, Qualcomm, and startups).

Current and former members include Rhodes Scholars, Clarendon Scholars, DeepMind Scholars, one cancer research UK Scholar, with our students funded by many additional sources (AIMS CDT, Cyber CDT, industry, and more). If you'd like to join us take a look here.

Current group members: Yarin Gal (Associate Professor) ● Freddie Kalaitzis (Senior Research Fellow, 2020) ● Tiarnan Doherty (Postdoc, 2022) ● Daniella (Zihuiwen) Ye (PhD, 2024) ● Xander Davies (PhD, 2024) ● Yoav Gelberg (PhD, 2024) ● Suhaas Bhat (Associate Member (PhD), 2024) ● Hazel Kim (PhD, 2023) ● Luckeciano Carvalho Melo (PhD, 2023) ● Matthew Kearney (PhD, 2023) ● Yonatan Gideoni (PhD, 2023) ● Katrina Dickson (Associate Member (Program Manager), 2023) ● Kunal Handa (MSc by Research, 2023) ● Ilia Shumailov (Associate Member (Senior Research Fellow), 2022) ● Gunshi Gupta (PhD, 2021) ● Kelsey Doerksen (PhD, 2021) ● Ruben Weitzman (PhD, 2021) ● Shreshth Malik (PhD, 2021) ● Lisa Schut (PhD, 2020) ● Muhammed Razzak (PhD, 2020) ● Gabriel Jones (Associate Member (PhD), 2020) ● Freddie Bickford Smith (Associate Member (PhD), 2020) ● Atılım Güneş Baydin (Associate Member (Faculty), 2020) ● Pascal Notin (Associate Member (Senior Research Fellow), 2019) ● Tom Rainforth (Associate Member (Faculty), 2019) ● Sebastian Farquhar (Associate Member (Senior Research Fellow), 2017)

Group Invited Talks

Recent invited talks:

11 Jun 2024, 12:00, Room 051	Marco Sertoli (Tokamak Energy). ML for Fusion Research
14 May 2024, 12:00, Lecture Theatre A	Albert Jiang (Cambridge/Mistral). Trustworthy Automation in Mathematics with Language Models
03 May 2024, 13:00, Lecture Theatre A	Jonathan Frankle (Databricks). Training Modern LLMs from Scratch
05 Mar 2024, 16:00, Lecture Theatre A	Nicolas Papernot (University of Toronto). Characterizing Machine Unlearning through Definitions and Implementations

Group Collaborators and Affiliates

Our group collaborators include:

And our group affiliates include: