We're happy to present the CBL Alumni talk series! 🕑4-5pm BST every 2nd Friday This week with Andrew Gordon Wilson: talks.cam.ac.uk/talk/index/158… Next up: Mark van der Wilk Shakir Mohamed David Duvenaud Finale Doshi-Velez Ferenc Huszár If you'd like to attend, please e-mail [email protected]

In this paper, we use this RKHS to construct an interdomain inducing variable that leads to GP basis functions c(.) that are approximately identical to the activation functions in neural nets. Left: standard RBF basis functions Right: our ReLU basis functions

Dan Roy Delip Rao e/σ I just did a retrospective analysis of 2014 NeurIPS ... There was no correlation between reviewer quality scores and paper's eventual impact. inverseprobability.com/talks/notes/th…

A high level presentation of the two papers that we presented at #AISTATS2022 on learning univariate and multivariate quantile functions for probabilistic forecasting featuring in #AmazonScience!

Our new ICML paper shows how to build more interpretable Gaussian-Process Generalized Additive Models (GP-GAMs). With Xiaoyu Lu and Alexis Boukouvalas. Here’s our GP-GAM on the SUSY dataset.

In January, the NFL’s Next Gen Stats announced its new QB passing score metric, which addressed the inconsistency across plays, games, weeks, and seasons found in previous scores. Learn how utilizing a Spliced Binned-Pareto distribution method led to the improved passing metric.

Finding yourself repeatedly training your ML model because you’ve collected more data? Not sure what to do with the hyperparameters? We investigated for you! Check out the paper arxiv.org/abs/2306.16916 with the wonderful Huibin Shen, François-Xavier Aubet, David Salinas and Aaron Klein

Excited for the poster session this afternoon at AutoML_conf! I'll be talking about hyperparameter optimisation for increasing data set sizes and learning between tasks. Work with Huibin Shen, François-Xavier Aubet, David Salinas and Aaron Klein #AutoML23

I am in Vienna for #ICML2024 this week! Let's catch up if you are there too.

Super happy to share our work AlphaProof, an AI agent that couples LLMs and AlphaZero. When combined with AlphaGeometry 2, it solves this year’s International Maths Olympiad problems at the level of a silver medalist! dpmd.ai/imo-silver

Gemini 2.0 Flash is here! 🎊 We're quite excited about this model. It's better than Gemini 1.5 Pro on most benchmarks, but comparable in speed and latency to Gemini 1.5 Flash. It has multilingual native audio output, native image output, native tool use, and a new multimodal

🚀🚀