Matteo Pirotta retweeted
How do we get RL agents to plan over long horizons without drowning in compounding errors?
Instead of reasoning over primitive, step-by-step actions, what if they could compose pre-trained policies across multiple timescales?
Poster #1117 Wed 8th at 2:30 pm #ICML2026
Matteo Pirotta retweeted
1/ My first PhD paper is out! 🎓
Title: Flow Matching in Feature Space for Stochastic World Modeling
tldr: we build stochastic world models directly in high-dimensional DINOv3 feature space, instead of relying on low-dimensional VAE latents.
Matteo Pirotta retweeted
Which representations are meaningful for control?
We're presenting TD-JEPA as an oral at ICLR🇧🇷: a zero-shot reinforcement learning algorithm using self-prediction (JEPA) to learn representations that are predictive of long-term, policy-dependent behavior. It works pretty well!🧵
Matteo Pirotta retweeted
My first PhD paper is out! 🎓
"What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?"
tl:dr: JEPA-WMs for robotics: learn dynamics on top of visual encoders, optimize actions towards goal 👇
w/ @JimmyTYYang1, Jean Ponce, @AdrienBardes, @ylecun
Matteo Pirotta retweeted
Meet BFM-Zero: A Promptable Humanoid Behavioral Foundation Model w/ Unsupervised RL👉 lecar-lab.github.io/BFM-Zero…
🧩ONE latent space for ALL tasks
⚡Zero-shot goal reaching, tracking, and reward optimization (any reward at test time), from ONE policy
🤖Natural recovery & transition
Exciting PhD position open at FAIR in Paris. We are looking for a candidate to join our team and contribute to advancing the field of AI, especially reinforcement learning.
Find more details and apply below. Feel free to reach out to me by email.
metacareers.com/jobs/1922660…
Matteo Pirotta retweeted
DINOv2 meets text at #CVPR 2025! Why choose between high-quality DINO features and CLIP-style vision-language alignment? Pick both with dino.txt 🦖📖
We align frozen DINOv2 features with text captions, obtaining both image-level and patch-level alignment at a minimal cost. [1/N]
Matteo Pirotta retweeted
Honored that our paper Temporal Difference Flows received the Best Paper Award at the #ICLR2025 World Models Workshop, and has also been accepted as a spotlight for #ICML2025! All made possible with the exceptional team @AIatMeta!
📄arxiv.org/abs/2503.09817
xitter.cf/JesseFarebro/status/19…
3) At the World Models workshop, I'll be giving an oral on a new approach to learning a generative model of successor states through flow matching / diffusion.
📍Peridot 201 & 206
📅Mon 28 Apr 5 PM - 5:30 PM
Check out the paper on arXiv: arxiv.org/abs/2503.09817 with a full tweet thread coming soon 🙂.
Excited to share our latest work at Meta FAIR: introducing Meta Motivo (metamotivo.metademolab.com/), a behavioral foundation model for controlling virtual physics-based humanoid agents.
The result? A model that can solve a wide range of unseen tasks — including reward optimization, goal reaching, and motion tracking — without any extra learning or task-specific fine-tuning, while being robust to unobserved environment perturbations.
We're thrilled to open-source the pre-trained model, APIs, and training code, so you can easily integrate it into your own projects. Plus, we've set up an interactive demo where you can see Meta Motivo in action.
Code: github.com/facebookresearch/…
I'm going to ICML next week!
We are looking for a Postdoc to join our reinforcement learning team at Meta FAIR. Our research spans different aspects of RL, with a particular focus on unsupervised/self-supervised learning to build foundation models for AI agents
How to get in touch: i) send me an email (I would like to chat with you at ICML!) ii) apply online through our website: metacareers.com/jobs/9923248….
Please forward this information to anyone in your circle who might be interested in joining our team.
I will be attending ICML 2023. If you are on the job market and interested in deep reinforcement learning and foundational models, please reach out to me. I would be happy to chat about opportunities related to our project.
Matteo Pirotta retweeted
🎉Exciting news! Applications for the Reinforcement Learning Summer School Barcelona 2023 are now open! Don't miss your chance to learn from top experts in the field! Applications close on March 27!
Application form: forms.gle/Ermo3BX82YJjSgyu8
#RLSS
Matteo Pirotta retweeted
#ICLR2023 accepted paper:
"Contextual Bandits with concave rewards, and application to fair ranking", by Virginie Do (@gini_do), E.D., M. Pirotta (@teopir), A. Lazaric (@alelazaric), & N. Usunier. Special kudos to Virginie Do. Checkout the preprint here arxiv.org/abs/2210.09957
Matteo Pirotta retweeted
Dear @aistats_conf reviewers, it seems that gmail is classifying all the AISTATS mail as spam or at least this is what happens to me. So, please check your spam folder and participate in the discussions.
Please retweet!
Today at #NeurIPS2022 we are presenting BanditSRL, a novel representation learning strategy that can be paired with any contextual bandit algorithm to improve its performance. BanditSRL works with NNs and achieves SOTA in a variety of problems @MetaAI arxiv.org/abs/2210.13083
Bonus point: BanditSRL is provably efficient and achieves constant regret under suitable realizability assumptions!
With Andrea Tirinzoni, @papinimat, @ahtouati, @alelazaric