Viacheslav Borovitskiy

@vabor112.bsky.social

Assistant Professor in Machine Learning @ University of Edinburgh. PhD in Mathematics. Ex ETH Zurich. See https://vab.im/vacancies/

My first paper with my first PhD student @pedrocvieira.bsky.social just landed on arXiv! ๐ŸŽ‰ โ€œDo Deep Ensembles Actually Capture Uncertainty in Graph Neural Networks?โ€ Spoiler alert: the answer is largely noโ€”at least, not much more than a single model does. ๐Ÿงต๐Ÿ‘‡ ๐Ÿ“„ arxiv.org/abs/2605.22593

Do Deep Ensembles Actually Capture Uncertainty in Graph Neural Networks?

While deep ensembles are widely considered to be the default method for uncertainty quantification in deep learning, their effectiveness for graph-structured data is often simply assumed based on succ...

arxiv.org

I am hiring a fully-funded #PhD in #ML to work at the University of Edinburgh on ๐ ๐ž๐จ๐ฆ๐ž๐ญ๐ซ๐ข๐œ ๐ฅ๐ž๐š๐ซ๐ง๐ข๐ง๐  and ๐ฎ๐ง๐œ๐ž๐ซ๐ญ๐š๐ข๐ง๐ญ๐ฒ ๐ช๐ฎ๐š๐ง๐ญ๐ข๐Ÿ๐ข๐œ๐š๐ญ๐ข๐จ๐ง. Application deadline: 31 Dec '25. Starts May/Sep '26. Details in the reply. Pls RT and share with anyone interested!

Bild

Presenting today at #ICLR2025! Poster session 1, 10:00-12:30, #427. Oral Session 2F 16:18-16:30. iclr.cc/virtual/2025...

Kacper Wyrwal@wyrwalkacper.bsky.social ยท last yr.

Excited to share our ICLR 2025 oral "Residual Deep Gaussian Processes on Manifolds"! With @vabor112.bsky.social & @arkrause.bsky.social, we introduce manifold-to-manifold GPs that can be composed together, generalising deep GPs to manifolds. Applications include wind prediction & Bayes opt! 1/n

Schematic illustration of a scalar-valued residual deep GP with L hidden layers. The last layer is a scalar-valued GP on the manifold. If it is not present, the model is manifold-valued. If it is replaced with a Gaussian vector field (GVF), the model is a vector field on the manifold.

Amazingly, Kacper did the bulk of the work for this ICLR oral as his undergrad thesis ๐Ÿ’ช

Kacper Wyrwal@wyrwalkacper.bsky.social ยท last yr.

Excited to share our ICLR 2025 oral "Residual Deep Gaussian Processes on Manifolds"! With @vabor112.bsky.social & @arkrause.bsky.social, we introduce manifold-to-manifold GPs that can be composed together, generalising deep GPs to manifolds. Applications include wind prediction & Bayes opt! 1/n

Schematic illustration of a scalar-valued residual deep GP with L hidden layers. The last layer is a scalar-valued GP on the manifold. If it is not present, the model is manifold-valued. If it is replaced with a Gaussian vector field (GVF), the model is a vector field on the manifold.

๐Ÿš€ ๐๐ก๐ƒ ๐Ž๐ฉ๐ฉ๐จ๐ซ๐ญ๐ฎ๐ง๐ข๐ญ๐ฒ ๐š๐ญ ๐ˆ๐ฆ๐ฉ๐ž๐ซ๐ข๐š๐ฅ Looking for my first PhD student ๐Ÿ”ฌ Project: ๐ˆ๐ฆ๐ฉ๐š๐œ๐ญ ๐จ๐Ÿ ๐ž๐ง๐ฏ๐ข๐ซ๐จ๐ง๐ฆ๐ž๐ง๐ญ๐š๐ฅ, ๐ฏ๐ข๐ซ๐š๐ฅ, ๐›๐ž๐ก๐š๐ฏ๐ข๐จ๐ฎ๐ซ๐š๐ฅ & ๐ฉ๐ฌ๐ฒ๐œ๐ก๐จ-๐ฌ๐จ๐œ๐ข๐š๐ฅ ๐ž๐ฑ๐ฉ๐จ๐ฌ๐ฎ๐ซ๐ž๐ฌ ๐จ๐ง ๐ก๐ž๐š๐ฅ๐ญ๐ก Joint with LSHTM & UKHSA See links below. ๐Ÿ“… ๐ƒ๐ž๐š๐๐ฅ๐ข๐ง๐ž: 7 March 2024 ๐Ÿ“ฉ Questions? DM me! #PhD #HealthEquity #ImperialCollege

Good news! GeometricKernels now supports the new ๐‘ฏ๐’๐’…๐’ˆ๐’†-๐’„๐’๐’Ž๐’‘๐’๐’”๐’Š๐’•๐’Š๐’๐’๐’‚๐’ ๐’Œ๐’†๐’“๐’๐’†๐’๐’” for flow-type data on graphs (thnx Maosheng Yang). Example notebook: geometric-kernels.github.io/GeometricKer.... For the theory behind, see arxiv.org/abs/2310.19450. Some details below.

Bild

Check out the workshop "Uncertainty in multivariate, non-Euclidean, and functional spaces: theory and practice" Where: Isaac Newton Institute, Cambridge When: 6-9 May 2025 Application deadline: 12 Jan 2025. Link in reply (contributed talks/posters are welcome) I will be there!

Bild