Profile avatar
marvinschmitt.bsky.social
🇪🇺 AI/ML, Member @ellis.eu 🤖 Generative NNs, ProbML, Uncertainty Quantification, Amortized Inference, Simulation Intelligence 🎓 PhD+MSc CS, MSc Psych 🏡 marvinschmitt.github.io ✨ On the job market, DMs open 📩
521 posts 5,935 followers 468 following
Regular Contributor
Active Commenter
Posts 16 Comments 34

🧠 Check out the classic examples from Bayesian Cognitive Modeling: A Practical Course (Lee & Wagenmakers, 2013), translated into step-by-step tutorials with BayesFlow! Interactive version: kucharssim.github.io/bayesflow-co... PDF: osf.io/preprints/ps...

submitted 18 hours ago • 0 comments

At the end of my PhD, I reflected on uncertainty quantification research, and what might change with chatbots and LLM agents. This was now accepted as position paper at @icmlconf.bsky.social. Some of those future topics are already picking up pace, so have an evening read ☕ arxiv.org/abs/2505.22655

submitted 2 days ago • 0 comments

Always found the connection between rectified flows and optimal transport a bit confusing. Glad to see some clarity. TL;DR: Rectified flows ≠ optimal transport. arxiv.org/abs/2505.19712

submitted 3 days ago • 0 comments

New Fundamentals…for innovation, collaboration, scientific discovery & more. At our pre-launch event on May 21, @samikaski.bsky.social outlined the ELLIS Institute mission to fundamentally re-work AI and ML to transform R&D, enhance human-AI teamwork & create exciting new science. 👇🧵

submitted 7 days ago • 1 comment

Great paper. I especially liked this quote: "All we can do is postulate (often blatantly wrong) theories and rigorously analyze their explanatory scope and limits."

submitted 8 days ago • 0 comments

Because it wasn't annoying enough that my domain was down for 35 days... Now a bot network farm automatically bought my old personal domain the second it got released. Odds are that they will (1) host explicit+spam content on the website and/or (2) try to sell it back to me for an ungodly price.

submitted 8 days ago • 1 comment

The more I use AI to teach myself new skills, the more convinced I am that this is one of THE best ways to learn (especially if you have to learn selectively and quickly). However, strategy matters here. I threw some thoughts on using AI for skilling here. www.linkedin.com/posts/tompca...

submitted 11 days ago • 1 comment

I spent the last years building LLM infra for large scale info extraction projects. Our org got decimated by today's layoffs, and I'm a free agent. If you need a remote scientist with very applied methods and coding skills, please pass along. Details 👇

submitted 16 days ago • 1 comment

Aleatoric and epistemic uncertainty are clear-cut concepts, right? ... right? 😵‍💫 In our new ICLR blogpost we let different schools of thought speak and contradict each other, and revisit chatbots where “the character of aleatory ‘transforms’ into epistemic” iclr-blogposts.github.io/2025/blog/re...

submitted 23 days ago • 1 comment

The extended version of this just got accepted at #UAI! Stay tuned for more updates and more work on Bayesian inference in low-precision regimes.

submitted 23 days ago • 1 comment

Clouded mind. (created with #midjourney)

submitted 24 days ago • 0 comments

If you’re into bluesy tunes and parody lyrics, check out our cover song! 🎶 There ain’t no cure for the PhD blues 🎶

submitted 24 days ago • 0 comments

Congratulations to the #AABI2025 Workshop Track Outstanding Paper Award recipients!

submitted 31 days ago • 0 comments

1/ If you are at ICLR / AABI / AISTATS, check out work from our lab and collaborators on *inference everywhere anytime all at once*! Go talk to my incredible PhD students @huangdaolang.bsky.social & @chengkunli.bsky.social + amazing collaborator Severi Rissanen. @univhelsinkics.bsky.social FCAI

submitted 34 days ago • 1 comment

My domain is down. In the meantime, you can find my website at: marvinschmitt.github.io

submitted 44 days ago • 0 comments

🆕 The ELLIS Unit London (@ucl.ac.uk) will host the first workshop uniting disparate subfields of post-Bayesian methods, including PAC Bayes, generalized Bayes, predictive resampling, Martingale posteriors, and online learning. 📍 London 🏴󠁧󠁢󠁥󠁮󠁧󠁿 📅 May 15-16 🔗 bit.ly/3GgwZ0G

submitted 49 days ago • 1 comment