iiinova — where your digital self keeps evolving

Reposted by Luca Ambrogioni

Nadine Dijkstra @nadinedijkstra.bsky.social · 13/05/2026

Researching visual imagery? Consider submitting to this new cross-journal special issue on visual imagery at Nature Communications, Communications Psychology, and Scientific Reports! We welcome a broad range of topics and methods - more info below 👇 www.nature.com/collections/...

04012

Luca Ambrogioni @lucamb.bsky.social · 23/04/2026

1/2) How do patterns form in diffusion models? Out-of-equilibrium phase transitions! Symmetry breaks → low-frequency modes destabilize → large-scale structure emerges. The peper offers a statistical field theory analysis of this process! Link: arxiv.org/abs/2603.20092

040

Reposted by Luca Ambrogioni

Robert Rosenbaum @robertrosenbaum.bsky.social · 03/09/2025

The University of Notre Dame is hiring 5 tenure or tenure-track professors in Neuroscience, including Computational Neuroscience, across 4 departments. Come join me at ND! Feel free to reach out with any questions. And please share! apply.interfolio.com/173031

apply.interfolio.com

Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio

03933

Luca Ambrogioni @lucamb.bsky.social · 02/09/2025

I am very happy to finally share something I have been working on and off for the past year: "The Information Dynamics of Generative Diffusion" This paper connects entropy production, divergence of vector fields and spontaneous symmetry breaking link: arxiv.org/abs/2508.19897

0213

Luca Ambrogioni @lucamb.bsky.social · 12/06/2025

Many when the number of steps in the puzzle is in the thousands and any error leads to a wrong solution

000

Luca Ambrogioni @lucamb.bsky.social · 11/06/2025

Have you ever asked your child to solve a simple puzzle in 60.000 easy steps?

100

Reposted by Luca Ambrogioni

Cian O'Donnell @cianodonnell.bsky.social · 11/06/2025

Students using AI to write their reports is like me going to the gym and getting a robot to lift my weights

25816

Luca Ambrogioni @lucamb.bsky.social · 16/05/2025

Generative decisions in diffusion models can be detected locally as symmetry breaking in the energy and globally as peaks in the conditional entropy rate. The both corresponds to a (local or global) suppression of the quadratic potential (Hessian trace).

070

Reposted by Luca Ambrogioni

Casper Kerrén @ckerren.bsky.social · 29/04/2025

🧠✨How do we rebuild our memories? In our new study, we show that hippocampal ripples kickstart a coordinated expansion of cortical activity that helps reconstruct past experiences. We recorded iEEG from patients during memory retrieval... and found something really cool 👇(thread)

516661

Luca Ambrogioni @lucamb.bsky.social · 03/05/2025

Why? You can just mute out politics and owner's antics and it becomes perfecly fine again

430

Luca Ambrogioni @lucamb.bsky.social · 02/05/2025

In continuous generative diffusion, the conditional entropy rate is the constant term that separates the score matching and the denoising score matching loss This can be directly interpreted as the information transfer (bit rate) from the state x_t and the final generation x_0.

0215

Luca Ambrogioni @lucamb.bsky.social · 30/04/2025

Decisions during generative diffusion are analogous to phase transitions in physics. They can be identified as peaks in the conditional entropy rate curve!

0103

Reposted by Luca Ambrogioni

Dan Levenstein @dlevenstein.bsky.social · 29/04/2025

I'd put these on the NeuroAI vision board: @tyrellturing.bsky.social's Deep learning framework www.nature.com/articles/s41... @tonyzador.bsky.social's Next-gen AI through neuroAI www.nature.com/articles/s41... @adriendoerig.bsky.social's Neuroconnectionist framework www.nature.com/articles/s41...

23410

Reposted by Luca Ambrogioni

Christian A. Naesseth @canaesseth.bsky.social · 30/04/2025

Very excited that our work (together with my PhD student @gbarto.bsky.social and our collaborator Dmitry Vetrov) was recognized with a Best Paper Award at #AABI2025! #ML #SDE #Diffusion #GenAI 🤖🧠

1182

Luca Ambrogioni @lucamb.bsky.social · 29/04/2025

Indeed. We are currently doing a lot of work on guidance, so we will likely try to use entropic time there as well soon

120

Luca Ambrogioni @lucamb.bsky.social · 29/04/2025

The largest we have tried so far is EDM2 XL on 512 ImageNet. It works very well there! We did not try with guidance so far

120

Luca Ambrogioni @lucamb.bsky.social · 29/04/2025

I am very happy to share our latest work on the information theory of generative diffusion: "Entropic Time Schedulers for Generative Diffusion Models" We find that the conditional entropy offers a natural data-dependent notion of time during generation Link: arxiv.org/abs/2504.13612

2255

Reposted by Luca Ambrogioni

Luca Ambrogioni @lucamb.bsky.social · 27/11/2024

Flow Matching in a nutshell.

1517

Reposted by Luca Ambrogioni

Arno Solin @arnosolin.bsky.social · 08/12/2024

I will be at #NeurIPS2024 in Vancouver. I’m looking for post-docs, and if you want to talk about post-doc opportunities, get in touch. 🤗 Here’s my current team at Aalto University: users.aalto.fi/~asolin/group/

0155

Reposted by Luca Ambrogioni

Ferenc Huszár @inference.vc · 06/12/2024

Can language models transcend the limitations of training data? We train LMs on a formal grammar, then prompt them OUTSIDE of this grammar. We find that LMs often extrapolate logical rules and apply them OOD, too. Proof of a useful inductive bias. Check it out at NeurIPS: nips.cc/virtual/2024...

nips.cc

NeurIPS Poster Rule Extrapolation in Language Modeling: A Study of Compositional Generalization on OOD PromptsNeurIPS 2024

71138

Reposted by Luca Ambrogioni

John Chodera @jchodera.bsky.social · 06/12/2024

Excited to speak at the ELLIS ML4Molecules Workshop 2024 in Berlin! moleculediscovery.github.io/workshop2024/

Photograph of Johannes Margraph and Günter Klambauer introducing the ELLIS ML4Molecules Workshop 2024 in Berlin at the Fritz-Haber Institute in Dahlem.

3464

Luca Ambrogioni @lucamb.bsky.social · 06/12/2024

Can we please stop sharing posts that legitimate murder? Please.

010

Reposted by Luca Ambrogioni

Petar Veličković @petar-v.bsky.social · 05/12/2024

Our team at Google DeepMind is hiring Student Researchers for 2025! 🧑‍🔬 Interested in understanding reasoning capabilities of neural networks from first principles? 🧑‍🎓 Currently studying for a BS/MS/PhD? 🧑‍💻 Have solid engineering and research skills? 🌟 We want to hear from you! Details in thread.

2585

Reposted by Luca Ambrogioni

Bao Pham @baopham.bsky.social · 05/12/2024

Diffusion models create beautiful novel images, but they can also memorize samples from the training set. How does this blending of features allow creating novel patterns? Our new work in Sci4DL workshop #neurips2024 shows that diffusion models behave like Dense Associative Memory networks.

On the left figure, it showcases the behavior of Hopfield models. Given a query (the initial point of energy descent), a Hopfield model will retrieve the closest memory (local minimum) to that query such that it minimizes the energy function. A perfect Hopfield model is able to store patterns in distinct minima (or buckets). In contrast, the right figure illustrates a bad Associative Memory system, where stored patterns share a distinctive bucket. This enables the creation of spurious patterns, which appear like mixture of stored patterns. Spurious patterns will have lower energy than the memories due to this overlapping.

1405

Luca Ambrogioni @lucamb.bsky.social · 04/12/2024

The naivete of these takes is always amusing They could be equally applied to human beings, and they would work as well

010

Luca Ambrogioni @lucamb.bsky.social · 04/12/2024

There are indeed cases in which obtaining an SDE equivalence isn't straightforward

010

Luca Ambrogioni @lucamb.bsky.social · 04/12/2024

I have always been saying that diffusion = flow matching. Is it supposed to be some sort of news now??

140

Luca Ambrogioni @lucamb.bsky.social · 03/12/2024

However, flow matching theory doesn't provide much guidance on how to do stochastic sampling It relies on the extra structure of diffusion

110

Luca Ambrogioni @lucamb.bsky.social · 03/12/2024

Disagree, religious literacy is important

000

Reposted by Luca Ambrogioni

Nicolas Beltran-Velez @velezbeltran.bsky.social · 02/12/2024

I am very excited to share our new Neurips 2024 paper + package, Treeffuser! 🌳 We combine gradient-boosted trees with diffusion models for fast, flexible probabilistic predictions and well-calibrated uncertainty. paper: arxiv.org/abs/2406.07658 repo: github.com/blei-lab/tre... 🧵(1/8)

Samples y | x from Treeffuser vs. true densities, for multiple values of x under three different scenarios. Treeffuser captures arbitrarily complex conditional distributions that vary with x.

415223

Reposted by Luca Ambrogioni

ruiqigao.bsky.social @ruiqigao.bsky.social · 02/12/2024

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.

625459

Reposted by Luca Ambrogioni

Christian A. Naesseth @canaesseth.bsky.social · 30/11/2024

I'm still cautiously optimistic that we'll find a way to leverage Bayesian ideas in "Modern" AI without retrofitting. However, I'm very much an agnostic when it comes the philosophy of uncertainty (Bayes vs frequentist vs imprecise etc.)

0131

Reposted by Luca Ambrogioni

Xiaoxuan @xiaoxuanlei.bsky.social · 28/11/2024

🌟 New Research Alert! 🌟 Excited to share our latest work (accepted to NeurIPS2024) on understanding working memory in multi-task RNN models using naturalistic stimuli!: with @takuito.bsky.social and @bashivan.bsky.social #tweeprint below:

34917

Reposted by Luca Ambrogioni

Gautam Kamath @gautamkamath.com · 27/11/2024

Asking the following earnestly: what is the strongest case for GANs standing the "test of time"? Are they important 10 years later in modern ML research? How have they influenced the way we think about generative models today?

18645

Reposted by Luca Ambrogioni

sbi - Simulation-based inference @sbi-devs.bsky.social · 27/11/2024

The sbi package is growing into a community project 🌍 To reflect this and the many algorithms, neural nets, and diagnostics that have been added since its initial release, we have written a new software paper 📝 Check it out, and reach out if you want to get involved: arxiv.org/abs/2411.17337

arxiv.org

sbi reloaded: a toolkit for simulation-based inference workflows

Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challeng...

16022

Reposted by Luca Ambrogioni

kyunghyuncho.bsky.social @kyunghyuncho.bsky.social · 27/11/2024

congratulations, @ian-goodfellow.bsky.social, for the test-of-time award at @neuripsconf.bsky.social! this award reminds me of how GAN started with this one email ian sent to the Mila (then Lisa) lab mailing list in May 2014. super insightful and amazing execution!

318727

Reposted by Luca Ambrogioni

Mathurin Massias @mathurinmassias.bsky.social · 27/11/2024

Anne Gagneux, Ségolène Martin, @quentinbertrand.bsky.social Remi Emonet and I wrote a tutorial blog post on flow matching: dl.heeere.com/conditional-... with lots of illustrations and intuition! We got this idea after their cool work on improving Plug and Play with FM: arxiv.org/abs/2410.02423

1235299

Reposted by Luca Ambrogioni

Alexander Terenin @avt.im · 21/11/2024

I made a starter pack for machine learning researchers working on Bayesian optimization and Gaussian processes! It's *very* sparse ATM since the migration is still in progress. Please reply to make me aware of people - potentially yourself - who should be added to it! go.bsky.app/QYMEQ52

135610

Luca Ambrogioni @lucamb.bsky.social · 27/11/2024

Interesting! I would definitely be happy to brainstorm. Feel free to reach out!

000

Luca Ambrogioni @lucamb.bsky.social · 27/11/2024

Yes. I used the improper kernel -|x| by accident and I found out that it gave very meaning full results even though it wasn't positive definite. It took years to work out the theory and discover the links with many older results in statistics

220

Luca Ambrogioni @lucamb.bsky.social · 27/11/2024

Thanks! It took a lot of work. The funny thing is that all started from a bug in my code Mybe I'll work on a follow up on Bayesian Optimization at some point

110

Luca Ambrogioni @lucamb.bsky.social · 27/11/2024

I mostly do diffusion nowadays. The paper I shared was a passion project

100

Reposted by Luca Ambrogioni

Aron ⇋ e⁻ @aronwalsh.github.io · 23/11/2024

Imagine that double Spiderman meme, but with generative AI pointing back at statistical mechanics. A neat paper in #JPCL on performing thermodynamic integration using denoising diffusion models #CompChem #CompChemSky pubs.acs.org/doi/10.1021/...

The graphical abstract of a paper in J Phys Chem Lett illustrating thermodynamic integration based on a denoising diffusion model

1466

Reposted by Luca Ambrogioni

Stannis Zhou @stanniszhou.bsky.social · 23/11/2024

Hello world! Excited to (re)share from X our new paper on "Diffusion Model Predictive Control" (D-MPC). Key idea: leverage diffusion models to learn a trajectory-level (not just single-step) world model to mitigate compounding errors when doing rollouts. arxiv.org/abs/2410.05364 🧵 1/4

arxiv.org

Diffusion Model Predictive Control

We propose Diffusion Model Predictive Control (D-MPC), a novel MPC approach that learns a multi-step action proposal and a multi-step dynamics model, both using diffusion models, and combines them for...

1483

Reposted by Luca Ambrogioni

Mathieu Alain @miniapeur.bsky.social · 26/11/2024

Topological & geometrical machine learning starter pack go.bsky.app/HnBdLJU

6387