I want to highlight another item in addition to Arno who is clearly a rising star if not a star already: A future rising star Daolang Huang was a runner-up in the race. We will certainly hear more of him later! (Sorry to induce pressure on you Daolang...)

ELLIS Institute Finland is launching with public (Opetus- ja kulttuuriministeriö) and private support (Peter Sarlin) 🤝 A significant boost for AI R&D and a signal that Finland is investing in AI to attract the best talents and fuel growth & expertise in Europe. fcai.fi/news/2024/10/2… ELLIS

Let me know if you are around there as well. 🦋

The amount of time I spend on the first paragraph of a new intro and still end up with something that's barely any different from the previous papers is frankly embarrassing.

Postdoc, research fellow and PhD positions: the call to join the Finnish Center for #ArtificialIntelligence is now open. Check out our research areas and supervisors and apply by Feb 2, 2025: fcai.fi/winter-2025-re…

One week to go for the deadline!

William Laplante, Matias Altamirano, Andrew Duncan, Jeremias Knoblauch, Fran\c{c}ois-Xavier Briol. [stat.CO]. Robust and Conjugate Spatio-Temporal Gaussian Processes. arxiv.org/abs/2502.02450

My collaborator Ayush Bharti will be presenting our recently accepted AISTATS paper on 'cost-aware simulation-based inference' at the next One World ABI Seminar on the 27th February. arXiv paper: arxiv.org/abs/2410.07930 Seminar series: warwick.ac.uk/fac/sci/statis…

Ever used kernel mean embeddings or MMD? Wished you had a closed-form expression so you wouldn't need to estimate these? See our new paper with Alex Gessner, Toni Karvonen and Maren Mahsereci, which comes with an associated Python package. arxiv.org/abs/2504.18830

New ICML 2025 paper: Robust Spatio-Temporal GP Regression! We propose a new GP method with linear-in-time cost that is provably robust to outliers. Unlike competitors, our method is fully conjugate and requires no expensive variational inference! arxiv.org/abs/2502.02450

New ICML 2025 paper! arxiv.org/abs/2505.20433 Kernel mean embeddings are a super useful tool for representing and comparing distributions through the maximum mean discrepancy (MMD). In this paper, we consider embeddings of distributions using kernel quantiles instead of means.

Today is the day of the pre-ICML event at UCL! Come check out the exciting work from academics, industry researchers, postdocs and PhD students from around London: sites.google.com/view/pre-icml-… UCL Statistical Science UCL CSML