Valentin De Bortoli
@vdebortoli.bsky.social
messing up with gaussians
Reposted by Valentin De Bortoli
Remember how, when the lockdowns started, every organization said "we only have two weeks of cash on hand and will shut down if we don't get assistance"? That's basically happening to every single lab and NGO right now, except for no actual reason.
January 28, 2025 at 11:11 AM
Remember how, when the lockdowns started, every organization said "we only have two weeks of cash on hand and will shut down if we don't get assistance"? That's basically happening to every single lab and NGO right now, except for no actual reason.
Reposted by Valentin De Bortoli
For the French-speaking audience, S. Mallat's courses at the College de France on Data generation in AI by transport and denoising have just started. I highly recommend them, as I've learned a lot from the overall vision of his courses.
Recordings are also available: www.youtube.com/watch?v=5zFh...
Recordings are also available: www.youtube.com/watch?v=5zFh...
Génération de données en IA par transport et débruitage (1) - Stéphane Mallat (2024-2025)
YouTube video by Mathématiques et informatique - Collège de France
www.youtube.com
January 20, 2025 at 5:49 PM
For the French-speaking audience, S. Mallat's courses at the College de France on Data generation in AI by transport and denoising have just started. I highly recommend them, as I've learned a lot from the overall vision of his courses.
Recordings are also available: www.youtube.com/watch?v=5zFh...
Recordings are also available: www.youtube.com/watch?v=5zFh...
Reposted by Valentin De Bortoli
Slides for a general introduction to the use of Optimal Transport methods in learning, with an emphasis on diffusion models, flow matching, training 2 layers neural networks and deep transformers. speakerdeck.com/gpeyre/optim...
January 15, 2025 at 7:08 PM
Slides for a general introduction to the use of Optimal Transport methods in learning, with an emphasis on diffusion models, flow matching, training 2 layers neural networks and deep transformers. speakerdeck.com/gpeyre/optim...
Reposted by Valentin De Bortoli
January 5, 2025 at 4:48 AM
Reposted by Valentin De Bortoli
I'm delighted to note that our paper InDI has been selected as one of two Outstanding Paper awardees by the Transactions on Machine Learning @tmlr-pub.bsky.social
We sincerely thank the expert reviewers, Action Editors, the Outstanding Paper Committee, and the Editors for this honor
1/3
We sincerely thank the expert reviewers, Action Editors, the Outstanding Paper Committee, and the Editors for this honor
1/3
December 19, 2024 at 11:41 PM
I'm delighted to note that our paper InDI has been selected as one of two Outstanding Paper awardees by the Transactions on Machine Learning @tmlr-pub.bsky.social
We sincerely thank the expert reviewers, Action Editors, the Outstanding Paper Committee, and the Editors for this honor
1/3
We sincerely thank the expert reviewers, Action Editors, the Outstanding Paper Committee, and the Editors for this honor
1/3
Reposted by Valentin De Bortoli
The slides of my NeurIPS lecture "From Diffusion Models to Schrödinger Bridges - Generative Modeling meets Optimal Transport" can be found here
drive.google.com/file/d/1eLa3...
drive.google.com/file/d/1eLa3...
BreimanLectureNeurIPS2024_Doucet.pdf
drive.google.com
December 15, 2024 at 6:40 PM
The slides of my NeurIPS lecture "From Diffusion Models to Schrödinger Bridges - Generative Modeling meets Optimal Transport" can be found here
drive.google.com/file/d/1eLa3...
drive.google.com/file/d/1eLa3...
Reposted by Valentin De Bortoli
After watching this beautiful keynote by @arnauddoucet.bsky.social , I *had* to give these Schrodinger bridges a try! Very interesting to be able to "straighten" a basic flow-matching approach. Super cool work by @vdebortoli.bsky.social & co-author!
December 14, 2024 at 11:57 AM
After watching this beautiful keynote by @arnauddoucet.bsky.social , I *had* to give these Schrodinger bridges a try! Very interesting to be able to "straighten" a basic flow-matching approach. Super cool work by @vdebortoli.bsky.social & co-author!
Reposted by Valentin De Bortoli
Speaking at this #NeurIPS2024 workshop on a new analytic theory of creativity in diffusion models that predicts what new images they will create and explains how these images are constructed as patch mosaics of the training data. Great work by @masonkamb.bsky.social
scienceofdlworkshop.github.io
scienceofdlworkshop.github.io
SciForDL'24
scienceofdlworkshop.github.io
December 14, 2024 at 5:01 PM
Speaking at this #NeurIPS2024 workshop on a new analytic theory of creativity in diffusion models that predicts what new images they will create and explains how these images are constructed as patch mosaics of the training data. Great work by @masonkamb.bsky.social
scienceofdlworkshop.github.io
scienceofdlworkshop.github.io
Reposted by Valentin De Bortoli
I've been getting a lot of questions about autoregression vs diffusion at #NeurIPS2024 this week! I'm speaking at the adaptive foundation models workshop at 9AM tomorrow (West Hall A), about what happens when we combine modalities and modelling paradigms.
adaptive-foundation-models.org
adaptive-foundation-models.org
NeurIPS 2024 Workshop on Adaptive Foundation Models
adaptive-foundation-models.org
December 14, 2024 at 4:02 AM
I've been getting a lot of questions about autoregression vs diffusion at #NeurIPS2024 this week! I'm speaking at the adaptive foundation models workshop at 9AM tomorrow (West Hall A), about what happens when we combine modalities and modelling paradigms.
adaptive-foundation-models.org
adaptive-foundation-models.org
Reposted by Valentin De Bortoli
Fantastic #neurips keynote by Arnaud Doucet! Really like this slide tracing back many of the modern flow-matching / stochastic interpolants ideas to a 1986 result by probabilist Istvan Gyongy describing how to "Markovianize" a diffusion process (eg. having coefficients depending on all the past)
December 13, 2024 at 3:45 PM
Fantastic #neurips keynote by Arnaud Doucet! Really like this slide tracing back many of the modern flow-matching / stochastic interpolants ideas to a 1986 result by probabilist Istvan Gyongy describing how to "Markovianize" a diffusion process (eg. having coefficients depending on all the past)
🔥You enjoyed @arnauddoucet.bsky.social talk but want even more Schrodinger Bridge? Come talk to me at our poster!
🔷Schrodinger Bridge Flow for Unpaired Data Translation
🔊 East Exhibit Hall A-C #2504
Work done with my amazing collaborators Ira Korshunova
Andriy Mnih and @arnauddoucet.bsky.social
🔷Schrodinger Bridge Flow for Unpaired Data Translation
🔊 East Exhibit Hall A-C #2504
Work done with my amazing collaborators Ira Korshunova
Andriy Mnih and @arnauddoucet.bsky.social
December 13, 2024 at 5:52 PM
🔥You enjoyed @arnauddoucet.bsky.social talk but want even more Schrodinger Bridge? Come talk to me at our poster!
🔷Schrodinger Bridge Flow for Unpaired Data Translation
🔊 East Exhibit Hall A-C #2504
Work done with my amazing collaborators Ira Korshunova
Andriy Mnih and @arnauddoucet.bsky.social
🔷Schrodinger Bridge Flow for Unpaired Data Translation
🔊 East Exhibit Hall A-C #2504
Work done with my amazing collaborators Ira Korshunova
Andriy Mnih and @arnauddoucet.bsky.social
Reposted by Valentin De Bortoli
When a bunch of diffusers sit down and talk shop, their flow cannot be matched😎
It's time for the #NeurIPS2024 diffusion circle!
🕒Join us at 3PM on Friday December 13. We'll meet near this thing, and venture out from there and find a good spot to sit. Tell your friends!
It's time for the #NeurIPS2024 diffusion circle!
🕒Join us at 3PM on Friday December 13. We'll meet near this thing, and venture out from there and find a good spot to sit. Tell your friends!
December 12, 2024 at 1:15 AM
When a bunch of diffusers sit down and talk shop, their flow cannot be matched😎
It's time for the #NeurIPS2024 diffusion circle!
🕒Join us at 3PM on Friday December 13. We'll meet near this thing, and venture out from there and find a good spot to sit. Tell your friends!
It's time for the #NeurIPS2024 diffusion circle!
🕒Join us at 3PM on Friday December 13. We'll meet near this thing, and venture out from there and find a good spot to sit. Tell your friends!
Reposted by Valentin De Bortoli
Have you ever wondered why diffusion models memorize and all initializations lead to the same training sample? As we show, this is because like in dynamic systems, the memorized sample acts as an attractor and a corresponding attraction basin is formed in the denoising trajectory.
December 4, 2024 at 9:03 PM
Have you ever wondered why diffusion models memorize and all initializations lead to the same training sample? As we show, this is because like in dynamic systems, the memorized sample acts as an attractor and a corresponding attraction basin is formed in the denoising trajectory.
Reposted by Valentin De Bortoli
Optimal transport, convolution, and averaging define interpolations between probability distributions. One can find vector fields advecting particles that match these interpolations. They are the Benamou-Brenier, flow-matching, and Dacorogna-Moser fields.
December 4, 2024 at 1:55 PM
Optimal transport, convolution, and averaging define interpolations between probability distributions. One can find vector fields advecting particles that match these interpolations. They are the Benamou-Brenier, flow-matching, and Dacorogna-Moser fields.
Reposted by Valentin De Bortoli
Iterated RF with conservative vector fields should get to OT, though training remains a challenge
arxiv.org/abs/2209.14577
arxiv.org/abs/2209.14577
Rectified Flow: A Marginal Preserving Approach to Optimal Transport
We present a flow-based approach to the optimal transport (OT) problem between two continuous distributions $π_0,π_1$ on $\mathbb{R}^d$, of minimizing a transport cost $\mathbb{E}[c(X_1-X_0)]$ in the ...
arxiv.org
December 4, 2024 at 7:02 AM
Iterated RF with conservative vector fields should get to OT, though training remains a challenge
arxiv.org/abs/2209.14577
arxiv.org/abs/2209.14577
Reposted by Valentin De Bortoli
Hellinger and Wasserstein are the two main geodesic distances on probability distributions. While both minimize the same energy, they differ in their interpolation methods: Hellinger focuses on density, whereas Wasserstein emphasizes position displacements.
December 3, 2024 at 5:16 PM
Hellinger and Wasserstein are the two main geodesic distances on probability distributions. While both minimize the same energy, they differ in their interpolation methods: Hellinger focuses on density, whereas Wasserstein emphasizes position displacements.
Reposted by Valentin De Bortoli
This is a really nice blogpost by
@RuiqiGao and team that I enjoyed being a part of. My favorite key learnings are:
- DDIM sampler == flow matching sampling
- (Not) straight?
- SD3 weighting (Esser, Rombach, et al) is very similar to the EDM weighting (Karras, et al).
👇
@RuiqiGao and team that I enjoyed being a part of. My favorite key learnings are:
- DDIM sampler == flow matching sampling
- (Not) straight?
- SD3 weighting (Esser, Rombach, et al) is very similar to the EDM weighting (Karras, et al).
👇
A common question nowadays: Which is better, diffusion or flow matching? 🤔
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
December 3, 2024 at 1:26 PM
This is a really nice blogpost by
@RuiqiGao and team that I enjoyed being a part of. My favorite key learnings are:
- DDIM sampler == flow matching sampling
- (Not) straight?
- SD3 weighting (Esser, Rombach, et al) is very similar to the EDM weighting (Karras, et al).
👇
@RuiqiGao and team that I enjoyed being a part of. My favorite key learnings are:
- DDIM sampler == flow matching sampling
- (Not) straight?
- SD3 weighting (Esser, Rombach, et al) is very similar to the EDM weighting (Karras, et al).
👇
Reposted by Valentin De Bortoli
New datasets from @polymathicai.bsky.social available on @hf.co will train AI models to think like scientists. Read more: www.simonsfoundation.org/2024/12/02/n... #science #AI #machinelearning
New Datasets Will Train AI Models To Think Like Scientists
New Datasets Will Train AI Models To Think Like Scientists on Simons Foundation
www.simonsfoundation.org
December 2, 2024 at 9:28 PM
New datasets from @polymathicai.bsky.social available on @hf.co will train AI models to think like scientists. Read more: www.simonsfoundation.org/2024/12/02/n... #science #AI #machinelearning
Reposted by Valentin De Bortoli
A common question nowadays: Which is better, diffusion or flow matching? 🤔
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
December 2, 2024 at 6:45 PM
A common question nowadays: Which is better, diffusion or flow matching? 🤔
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.
Reposted by Valentin De Bortoli
I wrote a summary of the main ingredients of the neat proof by Hugo Lavenant that diffusion models do not generally define optimal transport. github.com/mathematical...
November 30, 2024 at 8:35 AM
I wrote a summary of the main ingredients of the neat proof by Hugo Lavenant that diffusion models do not generally define optimal transport. github.com/mathematical...
Reposted by Valentin De Bortoli
Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!
November 17, 2024 at 4:20 PM
Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!