Ryan-Rhys Griffiths (@Ryan_

Ryan-Rhys Griffiths

202 posts

Ryan-Rhys Griffiths

@Ryan__Rhys

San Francisco

Joined July 2020

Pinned
Ryan-Rhys Griffiths
@Ryan__Rhys
Apr 5, 2023
My PhD thesis on applications of C̶h̶a̶t̶G̶P̶T̶ #GaussianProcesses in the natural sciences is now available online! arxiv.org/abs/2303.14291 A belated thanks to my examiners! @andrewwhite01 and @pl219_Cambridge #AI4Science #GaussianProcesses #BayesianOptimization
51K
Ryan-Rhys Griffiths
@Ryan__Rhys
Dec 14, 2023
Come join myself, @6ojaHa , @samuel_stanton_ , @mistergtom , and @arian_jamasb at *Poster 108* from 17:00-19:00 at #NeurIPS2023 where we will be presenting GAUCHE: A library for Gaussian Processes in Chemistry! Paper: neurips.cc/virtual/2023/p… Code: github.com/leojklarner/ga…
48K
Ryan-Rhys Griffiths
@Ryan__Rhys
Aug 28, 2022
Excited to announce that I have joined @Meta as a postdoctoral researcher working on the Adaptive Experimentation team with @eytan and @MaxBalandat !
Ryan-Rhys Griffiths
@Ryan__Rhys
Jul 9, 2020
A tutorial on Gaussian Process Regression for molecules using GPflow! medium.com/@ryangriff123/… @GPflowProject
Ryan-Rhys Griffiths
@Ryan__Rhys
Jun 9, 2021
Excited to present (and open-source) a new approach for high-dimensional Bayesian optimisation which achieves state-of-the-art performance on property-guided molecule generation. Paper: arxiv.org/pdf/2106.03609… Code: github.com/huawei-noah/no…
Ryan-Rhys Griffiths
@Ryan__Rhys
May 7, 2024
Delighted to join @SGRodriques and @andrewwhite01 @FutureHouseSF last week. It was a very difficult decision to turn down a faculty position for this opportunity but these are unprecedented times in #AI4Science and I strongly believe in FutureHouse's mission.
10K
Ryan-Rhys Griffiths
@Ryan__Rhys
Dec 12, 2020
Check out our contributed talk on FlowMO: A Gaussian Process Library for Molecular Property Prediction today at 14:50 GMT at the #ML4Molecules Workshop @NeurIPSConf with @Henrymossmoss slideslive.com/38942397/gauss… Tutorials on string and Tanimoto kernels: github.com/Ryan-Rhys/Flow…
Ryan-Rhys Griffiths
@Ryan__Rhys
Dec 9, 2022
New work on molecular discovery using multioutput Gaussian processes. Lab experiments led by @JakeyLG @RaymondThawani @FuchterMatt Paper: pubs.rsc.org/en/content/art… Tutorial available in the Gauche library! Paper: arxiv.org/abs/2212.04450 Tutorial: github.com/leojklarner/ga…
Data-driven discovery of molecular photoswitches with multioutput Gaussian processes
From pubs.rsc.org
Ryan-Rhys Griffiths
@Ryan__Rhys
Nov 17, 2020
"Our Submission is Secretly a Strong Accept and You Should Treat it Like One" So hard to resist this reply during the ICLR rebuttal phase 🤣
Ryan-Rhys Griffiths
@Ryan__Rhys
Apr 10, 2024
Proceedings now available for GAUCHE: A Library for Gaussian Processes in Chemistry at #NeurIPS2023! Calling for contributors to support the next wave of applications for GP regression and Bayesian optimisation in #AI4Science Poster: neurips.cc/virtual/2023/p… Code:
56K
Ryan-Rhys Griffiths
@Ryan__Rhys
Aug 20, 2022
Many thanks Haitham! Docs available here: leojklarner.github.io/gauche/ The team hopes to update the repo and paper in the coming months! @Henrymossmoss @sangttruong @pschwllr @YuanqiD @arian_jamasb @6ojaHa @a_bourached @Gregory47922793 @austinjtripp
Haitham Bou Ammar
@hbouammar
Aug 19, 2022
Love it 😍 just love it 😍 amazing work Of BO people !!! Gaussian process and BO for #Chemistry from @Ryan__Rhys et al. #AI #MachineLearning Pls help tag others I couldn’t find most 🥳
leojklarner.github.io
Documentation
GAUCHE Logo GAUCHE is a collaborative, open-source software library that aims to make state-of-the-art probabilistic modelling and black-box optimisation techniques more easily accessible to scient...
Ryan-Rhys Griffiths
@Ryan__Rhys
Apr 29, 2022
Many thanks for hosting me @pschwllr !
LIAC at EPFL
@SchwallerGroup
Apr 29, 2022
Yesterday, we had the great pleasure of hosting Ryan (@Ryan__Rhys) from @Cambridge_Uni/@Mila_Quebec at EPFL. Thanks for sharing your insights on Gaussian Processes, kernels (e.g. the Tanimoto kernel -> arxiv.org/abs/2010.01118), and Bayesian Optimisation with us! #compchem
Ryan-Rhys Griffiths
@Ryan__Rhys
Aug 14, 2020
Excited to share some recent work led by @ChengBingqing on a codebase for visualising datasets of molecules and materials! pubs.acs.org/doi/full/10.10…
Ryan-Rhys Griffiths
@Ryan__Rhys
Feb 2, 2023
What are the mathematical capabilities of #ChatGPT? TLDR: To pass a university math exam you may do better to copy from your peers! Paper: arxiv.org/abs/2301.13867 ChatGPT, however, shows strong potential as a tool for mathematicians. Reach out to share your own experiences!
4.2K