Martin Jørgensen

50 posts

Martin Jørgensen banner
Martin Jørgensen

Martin Jørgensen

@JorgensenMart

Copenhagen, Denmark Katılım Mart 2015
540 Takip Edilen468 Takipçiler
Sabitlenmiş Tweet
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
💣I am excited to advertise 'Bézier Gaussian Processes for Tall and Wide Data' which is to appear at #NeurIPS2022 The paper presents a novel approach to scalable GP regression through a structured control points, which can be seen as a type of 'inducing points'. More in thread.
Martin Jørgensen tweet media
English
4
20
172
0
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
@maosbot For some current noise I would recommend IDLES, and I would start with the album ‘Joy as an act of resistance’ 🤘
English
1
0
1
185
Frederik
Frederik@FrederikWarburg·
#PhDone! Last week I defended my PhD on Probabilistic 3D Reconstruction 🎉 It was exciting to present my work on uncertainty estimation in neural networks, 3D/4D reconstruction, and image retrieval. The thesis can be found here: frederikwarburg.github.io/publications.h…
Frederik tweet mediaFrederik tweet mediaFrederik tweet mediaFrederik tweet media
English
12
5
133
9.1K
Martin Jørgensen retweetledi
Luigi Acerbi
Luigi Acerbi@AcerbiLuigi·
1/📯I am hiring again! Postdoc or PhD position in Sample-Efficient Probabilistic Machine Learning @UnivHelsinkiCS with strong links to @FCAI_fi Please see blurb below, and full ad here: helsinki.fi/en/researchgro… Applications evaluated on a rolling basis. Please RT!
Luigi Acerbi tweet media
English
3
20
36
9K
Alison Pouplin (on Bluesky 🦋)
Alison Pouplin (on Bluesky 🦋)@a_ppln·
I feel immensely fortunate to have had you and @mmbronstein as part of my external committee. It was a pleasure to present some of my work and I always learn so much from our interactions. 🙂
Frank Nielsen@FrnkNlsn

🎉Congratulations Dr Pouplin! @a_ppln for your PhD on Riemannian/Finslerian geometries for Machine Learning Happy to serve as examiner with Prof. Bronstein @mmbronstein Check TMLR @TmlrPub paper "Identifying latent distances with Finslerian geometry" twitter.com/TmlrPub/status…

English
8
10
130
24.5K
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
@maosbot I like it. Sweater says early 2000’s Australian rock, face says more 2010’s French electronic, wallpaper.. you tell me?
English
0
0
2
369
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
@thjashin Same here! Was wondering if they gave out free registrations to all reviewers this year when I got the email? 🤷‍♂️
English
1
0
1
0
Martin Jørgensen retweetledi
Masaki Adachi
Masaki Adachi@masaki_adachi·
Delighted to advertise our paper at #NeurIPS2022 It presents a novel, cheap, but more diverse gradient-free sampler for parallelising GP active learning. It scales to large batch sizes and we apply it to Bayesian inference via Bayesian quadrature, e.g., battery analytics. 🧵1/n
Masaki Adachi tweet media
English
1
9
36
0
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
In the paper we have scaled to over a millions data points, and to input dimension 385; which inferred 6^385 control points with variational inference. Many more details in the paper: arxiv.org/abs/2209.00343 Also, a huge shout-out to coauthor @maosbot
English
0
2
13
0
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
With that assumption, computations are 'straightforward' in what we call the Bézier buttress. A buttress is an architectural construct that supports a building. Inference and prediction with the GP is now merely a 'forward-pass' in a source-sink graph.
Martin Jørgensen tweet media
English
1
0
2
0
Martin Jørgensen
Martin Jørgensen@JorgensenMart·
💣I am excited to advertise 'Bézier Gaussian Processes for Tall and Wide Data' which is to appear at #NeurIPS2022 The paper presents a novel approach to scalable GP regression through a structured control points, which can be seen as a type of 'inducing points'. More in thread.
Martin Jørgensen tweet media
English
4
20
172
0
Martin Jørgensen retweetledi
Michael Cohen
Michael Cohen@Michael05156007·
“But they don’t scale”. Gaussian processes regression takes O(n^3) time for most kernels, hugely limiting their adoption. Sparse methods enabling O(n) time just aren’t that strong. With our kernel, a GP runs in O(n log n) time and beats the classic Matérn kernel. #NeurIPS2022 1/8
Michael Cohen tweet media
English
6
18
243
0
Martin Jørgensen retweetledi
Peter R. Hansen
Peter R. Hansen@ProfPHansen·
It is with great sadness that I learn of the passing of Ole Eiler Barndorff-Nielsen. (March 18, 1935 - June 26, 2022). "He lived a long and positive life, full of love and mathematics" his obituary reads. An academic giant whom I am fortunate to have worked with. RIP Ole.
Peter R. Hansen tweet media
English
1
12
69
0

Keşfet

@UnivHelsinkiCS @FCAI_fi @maosbot @wanxingchen_ @nguyentienvu @FrederikWarburg @a_ppln @mmbronstein