Claire Vernade @claireve.bsky.app

@vernadec

now on BlueSky @claireve. I tweet on Machine Learning, mostly Bandits, RL and Learning Theory. Professor at UTN, Nuremberg.

Nuremberg, Germany cvernade.com Joined November 2015

997 Following 4K Followers 1.2K Posts

Latest posts 5

Pinned post

Claire Vernade @claireve.bsky.app @vernadec Nov 12, 2025

📣 #ICML tutorials: We want to know what *you* would like to learn. This year, Adam White and I are calling for nominations of topics and/or presenters.

Until December 7th, you can send us your suggestions, and we will use them to shape the program.

icml.cc/Conferences/20…

Claire Vernade @claireve.bsky.app @vernadec Nov 4

I have to say I am quite mind-blown by how well this idea works. The gains at training time are huge and the performance are SOTA. I jokingly call it a "DeepSeek moment" but I genuinely think there's some of that in our Gaussian Action Readouts. Cool thread @YoussefMMSaied

Youssef Saied Verified account @YoussefMMSaied Oct 30

1/8) Flow policies are Gaussian 😮

We replace their last layer with a Gaussian head fit on 0.02% of data:
• Similar task success
• Up to 2x faster inference
• Closed-form likelihoods

And behavior can still be diverse via diffusion in physical time.
(Xiaomi-Robotics-1, GR00T)

View quoted post on X

Claire Vernade @claireve.bsky.app @vernadec Oct 14

We are calling for workshops at AISTATS 2027 in beautiful Montreal! virtual.aistats.org/Conferences/20…
Call deadline on October 30th, standard 3-page proposal format, we are looking forward to receiving your great ideas for workshops.
@aistats_conf @qberthet please RT :)

Claire Vernade @claireve.bsky.app @vernadec May 20

I’m really excited about this new work on Imitation Learning. My student @sheebaelhamd made a great thread, check it out! w/ @LucaViano4 ,@CevherLIONS

Ziyad Sheebaelhamd Verified account @sheebaelhamd May 19

1/

Real-world data is messy. In robotics and LLMs, datasets are built by hundreds of humans with completely different, often conflicting preferences.
How do you do Imitation Learning when your "experts" have contrasting goals? Enter Multi-Objective Imitation: 🧵

View quoted post on X

Claire Vernade @claireve.bsky.app @vernadec Feb 20

There will be 3 workshops at AISTATS this year -> consider submitting your work and attending!

Quentin Berthet @qberthet Feb 18

🌴📊📈 - AISTATS 2026 workshops

For the first time, AISTATS (May 2-5 in Tangier, Morocco) will have a day of workshops!

Please submit your best work by Feb. 27th:
- calibration-workshop.github.io
- optimal-workshop26.github.io
- causcale.github.io

virtual.aistats.org/virtual/2026/e…
👇
View quoted post on X