Guy Moss

On my way from Munich to Grenoble 🚞 to co-lead a 3-day SBI tutorial + hackathon together with @danielged.bsky.social, organised by Pedro Rodriguez and @ugrenoblealpes.bsky.social.

Excited to meet researchers from across France, many bringing their own simulators 🚀

Scientific poster with dark background and two black holes illustrated in the center. The paper visualizes gravitational waves, and explains how parameter estimation is performed with DINGO. The standard DINGO model and the DINGO-T1 architectures are illustrated and results are shown. For example, it is possible to reanalyze the same event with different detector configurations with DINGO-T1, illustrated bz a corner plot.

1/ 🌀 New paper alert! We introduce Dingo-T1, a flexible transformer-based deep learning model for gravitational-wave (GW) data analysis. It adapts to different detector & frequency settings, improving inference efficiency and flexibility

🚀 #AI #MachineLearning #Physics #Astronomy #AcademicSky

Finally got the job ad—looking for 2 PhD students to start spring next year:

www.gao-unit.com/join-us/

If comp neuro, ML, and AI4Neuro is your thing, or you just nerd out over brain recordings, apply!

I'm at neurips. DM me here / on the conference app or email if you want to meet 🏖️🌮

I’m at NeurIPS in San Diego this week to present cool work on foundation models for SBI!
Most importantly, I’ll be around to meet people and discuss science. 👨‍🔬

Our group is at NeurIPS and EurIPS this year with four papers and one workshop poster. If you are either curious about SBI with autoML, with foundation models, or on function spaces or about differentiable simulators with Jaxley, have a look below 👇 1/11

I’m very grateful to my colleagues @leahsmuhle.bsky.social , @coschroeder.bsky.social, Reinhard Drews and @jakhmack.bsky.social for making this happen! Come find me at #Eurips2025 or reach out to learn more!
(9/9)

On a real geoscientific problem of inferring the surface accumulation and basal melting rates of Antarctic ice shelves, FNOPE achieves equivalent performance to previous approaches using 2 orders of magnitude fewer simulations!
(8/9)

FNOPE can be extended to estimate incredibly large parameter domains, involving >16k parameters on the Darcy flow problem.
(7/9)

Our flexible approach is amortized and can estimate posterior distributions given any discretization of the parameter and/or observation domains without any additional training.
(6/9)

FNOPE can estimate posterior distributions over 1000-dimensional parameter spaces using as few as 100 simulations on benchmark tasks.
(5/9)

This key insight leads to natural methodological extensions, such as data augmentation via masking the training data, using non-uniform Fast Fourier Transforms to work on non-uniform discretizations, and simultaneous estimation of additional vector-valued parameters.
(4/9)

Our method uses Fourier Neural Operators (FNOs) for Posterior Estimation (FNOPE). By training flow matching models with an FNO backbone, we can take into account the inductive biases of continuous parameters, forming a natural way to represent distributions over smooth functions!
(3/9)

In SBI we train generative models for posterior estimation using model simulations. However, when the parameters of interest are function-valued, we end up with very high-dimensional parameter spaces, requiring huge numbers of training simulations.
(2/9)

I’m super excited to present our new work in #Eurips2025 and #Neurips2025! We developed FNOPE: a new simulation-based inference (SBI) method which excels at inferring function-valued parameters!

Paper: openreview.net/forum?id=yB5...
Code: github.com/mackelab/fnope
(1/9)

MackeLab has grown! 🎉 Warm welcome to 5(!) brilliant and fun new PhD students / research scientists who joined our lab in the past year — we can’t wait to do great science and already have good times together! 🤖🧠 Meet them in the thread 👇 1/7

Simulation-Based Inference: A Practical Guide A central challenge in many areas of science and engineering is to identify model parameters that are consistent with prior knowledge and empirical data. Bayesian inference offers a principled framewo...

Simulation-based inference (SBI) has transformed parameter inference across a wide range of domains. To help practitioners get started and make the most of these methods, we joined forces with researchers from many institutions and wrote a practical guide to SBI.

📄 Paper: arxiv.org/abs/2508.12939

🎉 sbi participated in GSoC 2025 through @numfocus.bsky.social and it was a great success: our two students contributed major new features and substantial internal improvements: 🧵 👇

Congrats to Dr Michael Deistler @deismic.bsky.social, who defended his PhD!

Michael worked on "Machine Learning for Inference in Biophysical Neuroscience Simulations", focusing on simulation-based inference and differentiable simulation.

We wish him all the best for the next chapter! 👏🎓

The Macke lab is well-represented at the @bernsteinneuro.bsky.social conference in Frankfurt this year! We have lots of exciting new work to present with 7 posters (details👇) 1/9

a man wearing a white shirt and tie smiles in front of a window ALT: a man wearing a white shirt and tie smiles in front of a window

I've been waiting some years to make this joke and now it’s real:

I conned somebody into giving me a faculty job!

I’m starting as a W1 Tenure-Track Professor at Goethe University Frankfurt in a week (lol), in the Faculty of CS and Math

and I'm recruiting PhD students 🤗

From hackathon to release: sbi v0.25 is here! 🎉

What happens when dozens of SBI researchers and practitioners collaborate for a week? New inference methods, new documentation, lots of new embedding networks, a bridge to pyro and a bridge between flow matching and score-based methods 🤯

1/7 🧵

Simulation-Based Inference: A Practical Guide A central challenge in many areas of science and engineering is to identify model parameters that are consistent with prior knowledge and empirical data. Bayesian inference offers a principled framewo...

Looky Looky! 😍🥳👏
arxiv.org/abs/2508.12939
Super fun project, I ❤️ed coauthoring w/ @sbi-devs.bsky.social.
Great lead by @deismic.bsky.social & @janboelts.bsky.social. Contribs by many talented people @jakhmack.bsky.social. 🙏 to #BenjaminKurtMiller for the kickstart! @helmholtzai.bsky.social

New preprint: SBI with foundation models!
Tired of training or tuning your inference network, or waiting for your simulations to finish? Our method NPE-PF can help: It provides training-free simulation-based inference, achieving competitive performance with orders of magnitude fewer simulations! ⚡️

I have been genuinely amazed how well tabpfn works as a density estimator, and how helpful this is for SBI ... Great work by @vetterj.bsky.social, Manuel and @danielged.bsky.social!!

My first paper on simulation-based inference (SBI) as part of @mackelab.bsky.social!

Exciting work on adapting state-of-the-art foundation models for posterior estimation. Almost plug-and-play, and surprisingly effective.

Paper/code in thread below 🧵

Mapping the Composition of Antarctic Ice Shelves as a Metric for Their Susceptibility to Future Climate Change We categorize Antarctic ice shelves into two parts: local meteoric ice and continental meteoric ice Buttressed ice shelves composed primarily of local meteoric ice are identified as being particu...

New paper in Geophysical Research Letters led by Vjeran Višnjević mapping out ice shelf areas which are maintained by local precipitation only doi.org/10.1029/2024...

Have I been to Antarctica? No. But my colleagues have, and we can learn a lot from the data they collected! Really happy to share that our work is now published!

More great news from the SBI community! 🎉
Two projects have been accepted for Google Summer of Code under the NumFOCUS umbrella, bringing new methods and general improvements to sbi. Big thanks to @numfocus.bsky.social, GSoC and our future contributors!

A wide shot of approximately 30 individuals standing in a line, posing for a group photograph outdoors. The background shows a clear blue sky, trees, and a distant cityscape or hills.

Great news! Our March SBI hackathon in Tübingen was a huge success, with 40+ participants (30 onsite!). Expect significant updates soon: awesome new features & a revamped documentation you'll love! Huge thanks to our amazing SBI community! Release details coming soon. 🥁 🎉

🎓Hiring now! 🧠 Join us at the exciting intersection of ML and Neuroscience! #AI4science
We’re looking for PhDs, Postdocs and Scientific Programmers that want to use deep learning to build, optimize and study mechanistic models of neural computations. Full details: www.mackelab.org/jobs/ 1/5