David Rügamer
Associate Prof @ LMU Munich
PI @ Munich Center for Machine Learning
Ellis Member
Associate Fellow @ relAI
-----
davidruegamer.github.io | muniq.ai
-----
BNNs, UQ in DL, DL Theory (Overparam, Implicit Bias, Optim), Sparsity
- Reposted by David RügamerWhat is wild to me is the defense, BY THE NEURIPS BOARD, that fabricated citations do not mean "the content of the papers themselves [is] necessarily invalidated" It does. It very much does. What do you think citing other work is for? What do you think writing a paper is for? What do you *think*?
- NEW: NeurIPS,one of the world’s top academic AI conferences, accepted research papers with 100+ AI-hallucinated citations, new report claims fortune.com/2026/01/21/n...
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David RügamerNew blog post (on a shiny new ICML blog!): What's New in #ICML2026 Peer Review Some highlights: - Policies to combat thinly sliced contributions - Cascading desk rejections for peer-review abuse - Reviewer reciprocity - New ways to support authors and reviewers Post: blog.icml.cc/2026/01/08/w...
- Reposted by David Rügamer
I'd like to propose the following norm for peer review of papers. If a paper shows clear signs of LLM-generated errors that were not detected by the author, the paper should be immediately rejected. My reasoning: 1/ #ResearchIntegrity - EurIPS was pretty fun! Thanks to @lacerbi.bsky.social , @conorhassan.bsky.social , and all organizers of the Amortized ProbML team for the great workshop and for inviting me as a speaker! 🙌 Also great to be able to present our spotlight paper on sparsity on the main stage 😍
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- OpenReview's Statement Regarding API Security Incident - November 27, 2025
- OpenReview had a bug that allowed crawling reviewer names for ICLR (and apparently for all previous and ongoing conferences), and it had been known since Nov 12 ................
- Reposted by David Rügamer[Not loaded yet]
- Cross-posting from X 👇 (cc @iclr-conf.bsky.social — don’t forget about the butterflies!)
- Reposted by David RügamerMunich is now the #2 AI ecosystem in Europe (after Paris) and #11 globally, per Dealroom’s Global Tech Ecosystem Index 2025! 🇩🇪🤖🇫🇷 Proud of the growing strength of the Bavarian AI ecosystem 👉 baiosphere.org �� dealroom.co/tech-ecosystem-index-2025�5 #AI #Munich #baiosphere
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Works like a charm for structured sparsity too 🔥 Tested on a variety of architectures: sparse attention heads, sparse inputs, sparse conv filters, sparse NODEs,... And no setup changes needed, just a regularized gating variable! Spotlight @NeurIPS + Oral @EurIPS 🤩 Preprint: arxiv.org/abs/2509.23898
- Have you ever wondered how to incorporate (non-)convex L_{2/D}, D>=2, regularization into your neural network while still optimizing it with SGD --- despite the non-smoothness of the penalty? Check out our new paper accepted at ICLR 2025: arxiv.org/abs/2502.02496
- 🚀 RTG-METEOR is hiring! Apply now for Doctoral (4y) or Postdoc (3y) positions (TV-L E13, 100%) at LMU & TUM. Research focus: bridging Machine Learning & Control Theory for complex dynamical systems. 📅 Deadline: Nov 13, 2025 📍 Start: Apr 1, 2026 www.portal.graduatecenter.lmu.de/ocgc/meteor
- Reposted by David Rügamer[Not loaded yet]
- Some tweet-to-flap mapping: NeuReps (www.neurreps.org) deadline extended by a week: x.com/neur_reps/st...
- Reposted by David Rügamer[Not loaded yet]
- Can transformers learn full Bayesian inference in context? 🤔 👉 Come find out and visit our poster E-1205 in the East Hall this morning at #ICML2025 presented by @arikreuter.bsky.social icml.cc/virtual/2025...
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Two open postdoctoral positions in applied AI from my heritage science colleagues next door: > Postdoctoral Researchers in AI and Machine Learning Applications in Heritage Science and Archaeology 1) job-portal.lmu.de/jobposting/d... 2) job-portal.lmu.de/jobposting/1...
- Reposted by David Rügamer[Not loaded yet]
- Open AI for science PhD/Postdoc position at the Institute for Assyriology and Hittitology in collaboration with muniq.ai and the machine learning consulting unit of the MCML ( @munichcenterml.bsky.social ). Deadline: May 19 job-portal.lmu.de/jobposting/e...
- It seems that we have 3 accepted papers at ICML 2025 🔥
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David Rügamer[Not loaded yet]
- Arriving in Singapore this afternoon 🛬 I'll attend #ICLR2025, #AABI2025, and #AISTATS2025 together with many of my students and collaborators to present our 2 orals, 5 posters, and 14 workshop contributions 🚀 Feel free to drop by!
- Working on probabilistic modeling, inference, and decision-making? Join us at #AABI 2025 if you’re going to Singapore later this month! Register here (free but spots are limited!): approximateinference.org
- Still enough time to switch back to Vancouver for this year's @neuripsconf.bsky.social ? 😬
- French researcher, going to a conference in Houston, was forbidden entry to US; his work and personal electronics were both confiscated. Why? Because a “random search” of his cell phone revealed a negative personal opinion on Trump and the Trump administration. www.lemonde.fr/internationa...
- Connecting neural network solutions via a low-loss path reveals a subspace of highly performant hypotheses. In our AIStats paper (arxiv.org/pdf/2503.03382), we study direct optimization of such paths and their ambient spaces 🌐. Better understanding this space can enable improved subspace sampling.
- Interested in blazingly fast (10x faster than previous methods), effectively tuning-free, high-performance sampling-based inference for Bayesian Neural Networks? Then check out our ICLR 2025 paper on Microcanonical Langevin Ensembles (MILE)! 🔥 openreview.net/pdf?id=QMtrW...
- Our paper "Preventing Sensitive Information Leakage via Post-hoc Orthogonalization with Application to Chest Radiograph Embeddings" got accepted at PAKDD 2025 🔥 with Tobias Weber, Michael Ingrisch, Bernd Bischl
- Have you ever wondered how to incorporate (non-)convex L_{2/D}, D>=2, regularization into your neural network while still optimizing it with SGD --- despite the non-smoothness of the penalty? Check out our new paper accepted at ICLR 2025: arxiv.org/abs/2502.02496
- Reposted by David Rügamer[Not loaded yet]
- Reposted by David RügamerDon't forget #AABI, the Symposium on Advances in Approximate Bayesian Inference is coming to Singapore!! Co-located #ICLR2025 Workshop Track: February 7, AoE Proceedings Track: February 7, AoE Fast Track: February 18 / March 14, AoE approximateinference.org/call/ #ML #Bayes #GenAI
