Romain Tavenard

Romain Tavenard

@rtavenar.bsky.social

Prof. at Univ. Rennes 2 ML for time series + a bit of Optimal Transport Maintainer of tslearn rtavenar.github.io

172 followers129 following25 posts

Romain Tavenard reposted

Inverse optimization seeks to recover the parameters of an optimization problem from observed solutions. When the objective depends linearly on these parameters, the gap loss is convex, yielding a convex formulation. Inverse optimal transport is one example. www.gpeyre.com/blog/2026/10...

Topics: 1. Differentiable shortest paths on graphs with cycles 2. Sparse partial OT plans via slicing 3. PAC-Bayes guarantees for sliced OT 4. Differentiable time series segmentation 5. When are sliced OT plans exact? Reposts welcome 🙏 #AcademicJobs #PhD #MachineLearning

Romain Tavenard reposted

⌛Less than 5 days left to submit your research work! Deadline September 5, AoE (hard deadline, no extension) #NeurIPS #generalization #machinelearning #timeseries #greenai #privacy #iot #efficiency #edgecomputing #latency #inference

aboisbunon.bsky.social@aboisbunon.bsky.social

I'm happy to announce the #NeurIPS2026 workshop TS-LIMITS (Generalization for Time Series in Tight Settings: Latency, Inference, Memory, prIvacy and susTainability), which will be hosted in Paris, France 🇫🇷, Dec 12–13 2026! 🔗 Check out details and call for papers: ts-limits.github.io

Romain Tavenard reposted

I'm happy to announce the #NeurIPS2026 workshop TS-LIMITS (Generalization for Time Series in Tight Settings: Latency, Inference, Memory, prIvacy and susTainability), which will be hosted in Paris, France 🇫🇷, Dec 12–13 2026! 🔗 Check out details and call for papers: ts-limits.github.io

Romain Tavenard reposted

Romain Tavenard reposted

Bienvenue à Nice, ville avec une des plus basses températures de France ! (ps: je recrute potentiellement un.e postdoc sur 24 mois d'ici fin 2026 sur des questions, plutôt théoriques, liées aux LLMs en post-training / alignment, je dis ça, je dis rien)

This prize is a well deserved recognition for awesome work. If you're interested in domain adaptation, you should definitely have a look at Theo's phd thesis and pip install skada ! theses.hal.science/tel-05409110v1

theses.hal.scienceMaking sure you're not a bot!
Théo Gnassounou@tgnassou.bsky.social

I'm thrilled to have been awarded the thesis prize by the SSFAM 🥳 Thank you to the jury for the recognition you've given me 🙏 A big thanks to @rflamary.bsky.social and Alex to supporting me in this PhD! I'm looking forward to present my work at CAp 2026 in Montpellier, see you there :)

Romain Tavenard reposted

I'm thrilled to have been awarded the thesis prize by the SSFAM 🥳 Thank you to the jury for the recognition you've given me 🙏 A big thanks to @rflamary.bsky.social and Alex to supporting me in this PhD! I'm looking forward to present my work at CAp 2026 in Montpellier, see you there :)

Romain Tavenard reposted

🚀 Excited to announce the release of skada v0.6.0! This new release brings important additions to the library and continued work to keep the ecosystem compatible with Sklearn!

Romain Tavenard reposted

FAQ on NeurIPS Europe: NeurIPS Europe is an official NeurIPS 2026 satellite event taking place in Paris, France, alongside the main conference in Sydney and the other satellite event in Atlanta. NeurIPS authors can present their papers at any of the three locations, subject to space availability.

Romain Tavenard reposted

🎓 Doing a PhD in AI but unsure how to navigate your future career options? Join #ELIAS for Diverse Career Trajectories within Sustainable AI and AI for Sustainability. Speaker: Laetitia Chapel (IRISA 🇫🇷) 📍 Virtual Webinar 📅 20 May 🕒 15:00 - 16:30 CET 🔗 Register now at https://bit.ly/3P9CIuh

Romain Tavenard reposted

Now accepted at ICML 2026! Congrats to Magamed :)

Samuel Vaiter@samuelvaiter.com

🧵 New preprint! Towards Understanding Steering Strength w/ M. Taimeskhanov and D. Garreau Activation steering is a popular way to control LLM behavior at inference. But how much should you steer? We provide the first theoretical analysis of the steering strength α. 📄 arxiv.org/abs/2602.02712 1/7

Romain Tavenard reposted

My PhD thesis manuscript will be available in the coming months, but I’ve written two blog posts based on the related work chapter: 1. Generative modeling with flow-based models 🪚 2. Data-translation with flow and diffusion bridges 🔨 Open to feedback and discussions! lebellig.github.io/blog/

Romain Tavenard reposted

Thibaut Germain and Karim Lounici will present today at ICLR Poster Session 3 Pavilion 3 our paper introducing the Spectral-Grassmann OT that is a proper distance between linear operators dynamical systems (Koopman) seen as distributions. It can notably be used to compute barycenters of systems.

Romain Tavenard reposted

Today NeurIPS is announcing our official satellite event in Paris. After responding to the call from Ellis following the success of EurIPS in December, we are pleased to reach a new milestone by joining forces with the NeurIPS organizing committee for the 2026 edition.

Romain Tavenard reposted

🧵 New preprint! Towards Understanding Steering Strength w/ M. Taimeskhanov and D. Garreau Activation steering is a popular way to control LLM behavior at inference. But how much should you steer? We provide the first theoretical analysis of the steering strength α. 📄 arxiv.org/abs/2602.02712 1/7

Romain Tavenard reposted

Romain Tavenard reposted

Romain Tavenard reposted

Ami-es du #GRETSI, qq chiffres pour "GRETSI Info" : progressions (LinkedIn, chaîne YouTube), relative déception en libre (Mastodon). Pour faire mieux : partagez ce QR code ! 38 Bluesky 17 Mastodon 29 X 71 Youtube 186 LinkedIn page 294 LinkedIn contacts 354 LinkedIn abonnements

Romain Tavenard reposted

Openreview opened the door to continuous and major revisions that nobody has time to check properly. I think that we should come back to short one pdf page replies to reviews. It would mean having decisions quicker so that we actually have time to work on papers before resubmitting them.