iiinova — where your digital self keeps evolving

Julian Rodemann @jurodemann.bsky.social · 07/07/2026

At #ICML2026? Come by our poster: We derive generalization bounds when predictions change the sample, the population, or both... ... and uncover a trade-off between changing the world and learning from it 🧐 📄 openreview.net/forum?id=uQy... 📍 Wed, Jul 8, 2.30 PM -- Session 4

000

Reposted by Julian Rodemann

Krikamol Muandet @krikamol.bsky.social · 01/05/2026

What does generalization mean under performativity? To learn more, check out this paper from @jurodemann.bsky.social

011

Reposted by Julian Rodemann

Natalie Collina @ncollina.bsky.social · 09/04/2026

Excited to give a tutorial this year at ICML 2026!

1155

Julian Rodemann @jurodemann.bsky.social · 01/05/2026

🍾🍾Two papers accepted at #ICML @icmlconf.bsky.social One of them is already on arxiv: arxiv.org/abs/2602.04402 Check it out if you want to understand generalization/stat. inference under performativity! Great collaboration w @krikamol.bsky.social, James Bailie and Unai Fischer-Abaigar

arxiv.org

Performative Learning Theory

Performative predictions influence the very outcomes they aim to forecast. We study performative predictions that affect a sample (e.g., only existing users of an app) and/or the whole population (e.g...

040

Reposted by Julian Rodemann

Vincent Fortuin @vincefort.eurosky.social · 23/03/2026

Join us this year for NeurIPS in Paris! This time, it will be an *official presentation venue* I’m very proud of the whole team for pulling this off, especially on such a short timeline since EurIPS last December 💪

0284

Reposted by Julian Rodemann

Symposium on Probabilistic Machine Learning @probml.bsky.social · 10/02/2026

ProbML 2026 (formerly AABI) invites submissions on probabilistic ML (both Bayesian and otherwise!), July 5 in Seoul (co-located with ICML). Website: probml.cc. Tracks: proceedings (PMLR), workshop, fast track. New focus includes applications in healthcare and climate! Submit by: 20 March 2026.

12516

Julian Rodemann @jurodemann.bsky.social · 06/02/2026

Predictions can affect the thing they aim to predict. Think of self-fulfilling economic forecasts or self-negating routing (Google maps etc.). Can we learn from samples that reacted to our model? And what if the ground truth also reacts? 👇Answers👇 arxiv.org/abs/2602.04402 w/ @krikamol.bsky.social

arxiv.org

Performative Learning Theory

011

Julian Rodemann @jurodemann.bsky.social · 16/12/2025

🚨 Paper Alert: Empirical Decision Theory 🚨 We introduce “Empirical Decision Theory”, a framework for decision making that is radically empirical: Instead of positing an abstract state space, we work with observed act–consequence protocols only. Check it out! 👉https://arxiv.org/abs/2512.05677👈

010

Reposted by Julian Rodemann

let-all.com @let-all.com · 03/12/2025

Reminder that this is today at 7 pm! Please join us if you are at #NeurIPS2025

084

Julian Rodemann @jurodemann.bsky.social · 03/12/2025

👋 I’m at @neuripsconf.bsky.social in San Diego 🌴Hit me up if you’d like to talk imprecise probabilities, learning theory, good old statistics, or simply anything Bayesian 🤓 #NeurIPS2025 #NeurIPS #NeurIPS25

030

Julian Rodemann @jurodemann.bsky.social · 01/12/2025

I had a great time visiting the @Stanford Trustworthy AI lab (stairlab.stanford.edu) last week. My talk on empirical AI alignment (lnkd.in/d_SxW4YP) and incentive-aware AI regulation sparked a lot of fruitful discussion💡 Now off to #NeurIPS2025 in San Diego. 🌴

040

Reposted by Julian Rodemann

Clément Canonne @ccanonne.github.io · 29/11/2025

Something worrying me: many seem to change their research direction out of FOMO, reacting to (the obvious) recent trend: "If I don't do this, someone else will do it!" One of the key perks we have in academia is the freedom to set our own agenda.* If someone else can—and WILL—do it, why would you?

6486

Julian Rodemann @jurodemann.bsky.social · 24/07/2025

🎉 Honored to receive the IJAR Young Researcher Award (🥈) The award is given to researchers who demonstrate excellence at an early stage of their scientific careers, sponsored by the International Journal of Approximate Reasoning (IJAR). isipta25.sipta.org/awards

021

Julian Rodemann @jurodemann.bsky.social · 10/06/2025

Glad to share this paper was accepted at @ecmlpkdd.org ! We show interpreting BayesOpt helps personalize soft exosuits (see picture) Great collaboation w/ Fede Croppi, @giuseppe88.bsky.social et al. @lmumuenchen.bsky.social @munichcenterml.bsky.social @harvard.edu @wyssinstitute.bsky.social

101

Reposted by Julian Rodemann

Sébastien Destercke @sebastiendestercke.bsky.social · 07/05/2025

Getting coverage guarantees over functional surrogate models? This is what A. Gray and V. Gopakumar (they did the heavy lifting) have done using conformal predictions and zonotopes over reduced dimensions. It will be present at UAI 2025 (@auai.org), but a preview is here: arxiv.org/abs/2501.18426.

arxiv.org

Guaranteed confidence-band enclosures for PDE surrogates

We propose a method for obtaining statistically guaranteed confidence bands for functional machine learning techniques: surrogate models which map between function spaces, motivated by the need build ...

032

Reposted by Julian Rodemann

Sébastien Destercke @sebastiendestercke.bsky.social · 26/04/2025

In case you are interested in very recent advances of learning under imprecision with probability sets, you can check out the recent and fantastic SIPTA virtual talk by @krikamol.bsky.social, now on youtube: www.youtube.com/watch?v=gGPF...

youtube.com

SIPTA Seminar by Krikamol Muandet: Imprecise generalisation

YouTube video by Imprecise Probabilities Channel of SIPTA

032

Reposted by Julian Rodemann

BERD@NFDI @berd-nfdi.bsky.social · 27/03/2025

Helen Alber's DAGStat 2025 talk yesterday showed how LLMs streamline sentence classification and how to fix their mistakes. 💡 The solution? A two-step approach: LLMs pre-classify, experts refine, and Sim-SIMEX corrects errors, boosting efficiency and handling imbalanced data as well as MC-SIMEX.

031

Julian Rodemann @jurodemann.bsky.social · 12/12/2024

If you're attending #NeurIPS2024, don't miss our spotlight poster on multicriteria benchmarking TODAY 11am in West Ballroom A-D Talk: neurips.cc/virtual/2024... Paper: openreview.net/pdf?id=jXxvS... Where? West Ballroom A-D #6501 When? Thu 12 Dec 11 a.m. - 2 p.m. local time #neurips #neurips24

neurips.cc

NeurIPS Poster Statistical Multicriteria Benchmarking via the GSD-FrontNeurIPS 2024

040

Julian Rodemann @jurodemann.bsky.social · 12/12/2024

#MachineLearning is all about learning parameters from data, right? Well, not quite… 🤸‍♂️Actually, it sometimes works the other way around. 🤔Curious how? 👉Check out our poster on reciprocal learning: neurips.cc/virtual/2024... @neuripsconf.bsky.social #NeurIPS2024 #NeurIPS

010