Associate Professor in Machine Learning @AaltoUniversity. Enjoying statistical machine learning. @ELLISforEurope Scholar. @yaf_fi
Joined January 2014
- Tweets 771
- Following 315
- Followers 2,909
- Likes 1,012
201 Photos and videos
Pinned Tweet
4 May 2019
Fresh from the press: "Applied Stochastic Differential Equations" with @simosarkka and published by @CambridgeUP.
Physical books: cambridge.org/fi/academic/su…
Online PDF version: users.aalto.fi/~asolin/sde-b…
Codes: github.com/AaltoML/SDE
8
92
373
Arno Solin retweeted
New preprint update! "Softly Constrained Denoisers for Diffusion Models Applied to Differential Equations": my first PhD project and first work in diffusion models. A thread on what we did and why: 🧵
1/n
5
2
10
613
Arno Solin retweeted
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
139
921
6,556
1,088,750
May 5
6/ ... and everyone who helped make the conference run so smoothly. It takes a lot of work behind the scenes to create a meeting like this.
1
1
6
2,076
May 5
7/ Aaditya and I now move on to serve as General Chairs for next year. We are very happy to welcome @qberthet and Aymeric Dieuleveut as the next Program Chairs. See you all in #Montréal for #AISTATS2027!
3
196
May 4
1
1
14
1,204
May 4
4/ On Day 3, Taiji Suzuki gave a keynote covering learning in and understanding transformers & diffusion models. His work on statistical learning theory, deep learning theory, kernel methods, and optimization has had a major influence on how we understand modern learning systems
1
6
48
4,100
May 4
5/ A huge thank you to all three for their inspiring talks and for contributing to the scientific program of @aistats_conf 2026!
2
333
May 3
1/ 🏆 Congratulations to the paper award winners for @aistats_conf #AISTATS2026! 👏👏👏
1
2
34
2,530
May 3
4/ 🏆 Test of Time Award: "Non-stochastic Best Arm Identification and Hyperparameter Optimization" by Kevin Jamieson and Ameet Talwalkar
1
1
6
2,056
May 3
5/ Additionally, we are awarding a 🏆 Test of Time, Honorable Mention to "Deep Kernel Learning" by Andrew Gordon Wilson, Zhiting Hu, Ruslan Salakhutdinov, and Eric P. Xing
2
22
4,999
Apr 30
.@massiviola01 doing a better job advertising our work than we do 😄 This is really interesting work, and credits to @wszhao_robot who came up with the idea.
Training a diffusion model has always been synonymous with one idea: add some noise to an image, then learn to remove it.
Since we know where we started and where we ended up, the natural thing to do is to ask the model to recover the signal everywhere.
But is it really needed?
2
10
4,250