Victor Letzelter

Research Scientist @ Mistral

[Victor Letzelter @VLetzelter]

@valeoai and @telecomparis are incredible places to do a PhD. Very grateful to the exceptional teams and to my advisors @abursuc @Mickael_Chen @ptrkprz @gillespuy @MathieuFontai19 @ccrommel Gaël and Slim

Many thanks to the jury @francoisfleuret @rakotal1 @irinarish @RFlamary

[valeo.ai @valeoai]

Congrats to our @VLetzelter for earning his well deserved PhD with his works on "Multiple Choice Learning from Ambiguous Signals".
Victor is a resourceful, curious and kind researcher. Good luck for next!

Fruit of an excellent collaboration between @telecomparis & @valeoai

[valeo.ai @valeoai]

Congrats to our @VLetzelter for earning his well deserved PhD with his works on "Multiple Choice Learning from Ambiguous Signals".
Victor is a resourceful, curious and kind researcher. Good luck for next!

Fruit of an excellent collaboration between @telecomparis & @valeoai

[Nicolas Sereyjol-Garros @sg_nicolas]

1/ 📢 New preprint: Test-Time Conditioning with Representation-Aligned Visual Features
Introducing REPA-G — a framework for controllable image generation at inference time using aligned visual features.
📄 https://t.co/bVlx3GqON5
💻 https://t.co/zIGVH25p3V
🧵👇

[Victor Letzelter @VLetzelter]

Let’s catch up if you’re at @icmlconf and want to chat!

[Adrien Crt @crtadrien1]

Interested in time series forecasting and data uncertainty quantification?
Check out our latest paper with @VLetzelter at @icmlconf !

Paper: https://t.co/4YOHCe7YZZ
Code: https://t.co/phdIpgEFQG
Poster #2211 , Tue 15 Jul 11 a.m. PDT East
#timeseries #quantization #uncertainty

[valeo.ai @valeoai]

🚗 Ever wondered if an AI model could learn to drive just by watching YouTube? 🎥👀

We trained a 1.2B parameter model on 1,800+ hours of raw driving videos.

No labels. No maps. Just pure observation.

And it works! 🤯

🧵👇 [1/10]

[Cédric Rommel 🦋 @ICLR 🇧🇷 @ccrommel]

Inferring 3D human poses from video is highly ill-posed because of depth ambiguity.

Our work accepted to #NeurIPS2024, ManiPose, gets one step closer to solving this, by leveraging prior knowledge about poses topology and cool multiple-choice learning techniques.

[valeo.ai @valeoai]

🌟 Calling all MSc students passionate about computer vision and ML!

We’re offering research internships about diffusion models, multi-modal transformers, continual learning, & more. 4 exciting openings await!

🔗 Learn more: https://t.co/HNVWpI5QJ3

RT to spread the word! 🙌

[Victor Letzelter @VLetzelter]

Check out the paper!

Arxiv: https://t.co/caolTDEqn2
Project page: https://t.co/gb7wSBiNmH
GitHub: https://t.co/JoqsHUZem5

A joint work with David Perera, Théo Mariotte, Adrien Cortes, @Mickael_Chen , Slim Essid and Gaël Richard at @valeoai and Télécom Paris.

[Victor Letzelter @VLetzelter]

Working on ill-posed machine learning tasks, interested in multi-heads neural networks and data #uncertainty quantification ?

Sharing here our latest research, which will be presented at @NeurIPSConf in December.

[Victor Letzelter @VLetzelter]

Key insights:
* Annealing enhances exploration compared to greedy convergence.
* Inspired by statistical physics and information theory, we describe the training trajectory.
* Experiments on synthetic datasets, UCI benchmarks, and speech separation show highly promising results.

[Björn Michele @Bjoern_Michele]

You have an already trained sem. segmentation model ?
You want to apply it to data with a domain shift ?
You are afraid of degradation during the adaptation ?

Then you might want to check out our work TTYD at @eccvconf 2024 in Milan.

👉 Poster # 73: Tuesday, 16:30
#ECCV2024

[Alain Riou @howariou]

Glad to announce that Stem-JEPA has been accepted to @ISMIRConf !

In this work, we tackle the task of musical stem compatibility estimation (what “fits” together) as a representation learning problem. (1/7)

Paper: https://t.co/Gx0NenVZes
Code: https://t.co/qpwzIdVN5C

[valeo.ai @valeoai]

.@VLetzelter will be at #ICML2024 to present his work on leveraging the geometric properties of the Winner-takes-all learners for conditional density estimation & uncertainty prediction, w/o modifying its original training scheme.
Find a tl;dr below & come say hi at the posters

[Victor Letzelter @VLetzelter]

Check out our paper at https://t.co/NWNkIJEoL9

A joint work with D.Perera, @ccrommel, @MathieuFontai19, S.Essid, G.Richard and @ptrkprz

Multi-head #neuralnetworks #machinelearning #Uncertainty #DensityEstimation

[Victor Letzelter @VLetzelter]

Interested in ill-posed learning tasks, uncertainty prediction, conditional density estimation or multi-head deep neural networks ?

In our new paper, accepted at #ICML24, we tackle these challenges by exploring the Winner-Takes-All (WTA) training scheme.
[1/n]

[Victor Letzelter @VLetzelter]

The attached figure illustrates the predictions (shaded blue points) made by WTA-based models compared to other baselines, for the task of estimating conditional distributions on a synthetic datasets (represented by green points).
[4/n]