←
Tom Rainforth376 posts
Tom Rainforth avatar

Tom Rainforth

@tom_rainforth

Professor of Statistical Machine Learning at the University of Oxford, Head of RainML Research Lab (rainml.uk)

◷ Joined November 2016▧ 48 Photos and videos
Tom Rainforth avatar
Tom Rainforth@tom_rainforth·1 Mar 2023

We've got really good at utilizing data. But methods for acquiring that data are often still rudimentary. Our new review paper shows how Bayesian experimental design has recently transformed to now provide a powerful mechanism to acquire data intelligently arxiv.org/abs/2302.14545

Post attachment
Tom Rainforth avatar
Tom Rainforth@tom_rainforth·Nov 5

I'm delighted to announce the release of TestBED, an open-source toolkit for Bayesian experimental design. It provides a flexible and extendable foundation for coding adaptive design methods and quickly benchmarking them arxiv.org/abs/2611.06633 github.com/testbed-benchmark…

Tom Rainforth avatar
Tom Rainforth@tom_rainforth·Aug 13

I'm delighted to share that I have been made a full professor at Oxford (specifically, I am now Professor of Statistical Machine Learning)

Kia avatar
⟳ Tom Rainforth retweeted
Kia@kiaashour·Jul 9

1/n 🧵 Model evaluation is often bottlenecked by labelling cost. Active testing estimates a model's risk from far fewer labels, yet existing methods overlook a now-common resource: cheap predictions from powerful black-box models. We address this with PPAT.

Post attachment
Jackson Atkins avatar
⟳ Tom Rainforth retweeted
Jackson Atkins@JacksonAtkinsX·1 Sep 2025

Apple and Oxford just made AI 6.5x better at problem-solving. The secret: it teaches AI agents to ask perfect questions. This rockets success rates from 14% to 91%. No need for fine-tuning or retraining. It runs on current models. Here's how it works: It's a strategic loop designed for multi-turn conversations. At every step, the agent works to find the shortest path to the right answer. Hypothesize: The agent creates an internal list of all possible solutions to the problem. Score Questions: It simulates asking various questions and scores each one on "Expected Information Gain" (EIG). This number represents how much a question is mathematically likely to shrink the list of possibilities. Ask the Best Question: It asks the user only the single, highest-scoring question. Update & Repeat: Based on the answer, it filters its list of hypotheses, getting smarter with each interaction, and then begins the loop again for the next turn. Why this matters for your AI strategy: This marks a shift from building passive "oracles" to proactive, question-asking agents Business Leaders: A 6.5x multiplier on task success is a lever for efficiency. This translates to fewer failed customer interactions, faster diagnostics, and more accurate personalization, a clear ROI on smarter AI. Practitioners: This is a deployment-time framework, not a new model. You can build this agent on top of existing LLMs today. It provides a principled way to overcome common multi-turn issues like inconsistency and context loss without fine-tuning or retraining. Researchers: This paper is a victory for information theory. It proves that a full EIG calculation is superior to heuristics like predictive entropy. It sets a new standard for how to build intelligent information-seeking agents.

Post attachment
Tom Rainforth avatar
Tom Rainforth@tom_rainforth·7 Jun 2024

I have an opening for a 2.5-year postdoc position in the RainML lab as part of my ERC grant on probabilistic machine learning and intelligent data acquisition. Application deadline 10th July 2024. See here for details and to apply: tinyurl.com/rainmlpostdoc

Tom Rainforth avatar
Tom Rainforth@tom_rainforth·7 Jun 2024

I'm delighted to announce that from September I will officially be an Associate Professor (remaining at the Oxford stats department)

Jannik Kossen avatar
⟳ Tom Rainforth retweeted
Jannik Kossen@janundnik·7 May 2024

Are you at ICLR? Have you heard that In-Context Learning in LLMs does not learn label relationships? Well that's not true. Visit our poster TODAY to find out how LLMs incorporate label information. Spoiler: it's not Bayesian inference. Poster #129, May 7, 4.30 pm

Post attachment
Tim Reichelt avatar
⟳ Tom Rainforth retweeted
Tim Reichelt@TimReichelt3·1 May 2024

I will be presenting our work on "Beyond Bayesian Model Averaging over Paths in Probabilistic Programs with Stochastic Support" at AISTATS in Valencia tomorrow (details in thread below). If you are interested in probabilistic programming, come and say hi at poster session 1!

Our new paper (arxiv.org/abs/2310.14888) shows that probabilistic programs with stochastic support are implicitly Bayesian model averages (BMA) which leads to issues if we assume our model is misspecified! w/ @TimReichelt3 and Luke Ong A thread (1/5)

Freddie Bickford Smith avatar
⟳ Tom Rainforth retweeted
Freddie Bickford Smith@fbickfordsmith·1 May 2024

The current default recipe for Bayesian active learning doesn’t really work beyond MNIST scale. We suggest why that is and identify a simple fix. arxiv.org/abs/2404.17249 @aistats_conf with @adamefoster @tom_rainforth 1/5