Jan Peters
@Jan_R_Peters
Followers
5K
Following
10K
Media
52
Statuses
1K
#RobotLearning Professor (#MachineLearning #Robotics) at @ias_tudarmstadt of @TUDarmstadt (Part of @ELLISforEurope, @Hessian_AI and @DFKI)
Darmstadt, Hesse, Germany
Joined July 2019
That's two days ago ... see where the @ias_tudarmstadt team (Kai and @_mlutter) are at NOW!!! https://t.co/N1UKKtLJG8 Talk about amazingly fast progress!
To continue the robot juggling! Now the second arm is also moving! 🤖☺️
3
11
76
Warming up for #IROS2025 🔥 We’re releasing CaDeLaC as open source! - Model and training pipeline - Integration with acados for adaptive physics-consistent MPC - Simulation + real Franka torque control 💻 https://t.co/w5AiWWwRWK
@ias_tudarmstadt @Jan_R_Peters @OlegArenz 🧵👇
github.com
Context-Aware Deep Lagrangian Networks for Model Predictive Control (CaDeLaC) - Schulze18/cadelac
1
3
7
⚡️ Can one unified policy control 10 million different robots and zero-shot transfer to completely unseen robots, even humanoids? 🔗 Yes! Checkout our paper: https://t.co/fUikut5ZNx If you are interested in massive multi-embodiment learning, come and chat with me at: - Today,
28
92
535
🚀 New preprint! Introducing XQC— a simple, well-conditioned actor-critic that achieves SOTA sample efficiency in #RL ✅ ~4.5× fewer parameters than SimbaV2 ✅ Scales to vision-based RL 👉 https://t.co/3xgo4tpbAK Thanks to Florian Vogt @JoeMWatson @Jan_R_Peters
@Hessian_AI @DFKI
1
4
12
We presented our work on studying multi-embodiment scaling at #CoRL2025 and were thrilled by the excitement around generalist cross-embodiment policies. A common question was: Will you move beyond locomotion? The answer is YES! Locomotion provides a clean starting point, but our
🧠 Can a single robot policy control many, even unseen, robot bodies? We scaled training to 1000+ embodiments and found: More training bodies → better generalization to unseen ones. We call it: Embodiment Scaling Laws. A new axis for scaling. 🔗 https://t.co/Gbebt7g494 🧵👇
0
13
48
Dear @davide_tateo, thank you for all the amazing work that you have done at our lab! All the best for our next step as faculty member at @lunduniversity !!!
Goodbye Robot Air Hockey! Goodbye Darias! Thank you so much @Jan_R_Peters for giving me the opportunity to work daily on such amazing platforms for the last six years 😁
0
0
17
As usual, #EWRL was a wonderful experience. I had the pleasure of presenting my research as a Contributed Talk 🎉 Special thanks to the organizers for making it happen!
1
1
13
Beyond excited to share that CrossQ+WN has been accepted at #NeurIPS2025
@Hessian_AI @ias_tudarmstadt @CS_TUDarmstadt @DFKI @NeurIPSConf
🚀 New preprint "Scaling Off-Policy Reinforcement Learning with Batch and Weight Normalization"🤖 We propose CrossQ+WN, a simple yet powerful off-policy RL for more sample-efficiency and scalability to higher update-to-data ratios. 🧵 https://t.co/Z6QrMxZaPY
#RL @ias_tudarmstadt
1
3
34
If you are interested, take a look at our paper ( https://t.co/UXNTrYNy3V) and don't hesitate to contact me if you have any questions! Huge thanks to my collaborators Guillaume Duret, @cristianaFar, @RCalandra, @limingchen69, and @Jan_R_Peters. @ias_tudarmstadt @CS_TUDarmstadt
1
1
6
⚡️ Want to build robots that are smarter, faster & truly agile? 🤝 Dreaming of seamless human–robot collaboration in shared spaces? 📢 Share your vision at the IROS 2025 Workshop on Robot Learning, Planning & Control in Dynamic Environments! 🌍 With speakers from MIT, Monash,
0
2
4
Today, @dyumanaditya presented our paper "Robust Localization, Mapping, and Navigation for Quadruped Robots" at ECMR 2025. It's always a pleasure to see that students and interns of @ias_tudarmstadt succeed and disseminate their ideas and results at conferences!
Do you want to deploy a navigation stack on a cheap quadruped robot? Take a look at our latest ECMR paper, "Robust Localization, Mapping, and Navigation for Quadruped Robots". In our work, we tackle the issues arising from the combination of low-cost robots and RL policies
1
4
20
And it's always great to collaborate with amazing people such as @piotr_kicki, @KrzysztofTWalas, and @goldleaf3. PS: If you are looking for a great supervisor (@Jan_R_Peters ) and a great team, apply to @ias_tudarmstadt, we have open positions!
0
1
6
Good job, @dymanaditya !!!
It was a pleasure to present our work on behalf of the @ias_tudarmstadt at #ecmr2025! It wouldn’t have been possible without the amazing supervisors: @davide_tateo, @goldleaf3, @Jan_R_Peters, @KrzysztofTWalas And amazing co-authors: @WangNing6, @NicoBohlinger, @piotr_kicki
0
0
5
⚡️ Want to build robots that are smarter, faster & truly agile? 🤝 Dreaming of seamless human–robot collaboration in shared spaces? 📢 Share your vision at the IROS 2025 Workshop on Robot Learning, Planning & Control in Dynamic Environments! 🌍 With speakers from MIT, Monash,
0
6
21
After so many requests, I want to update everyone on the status of Ark, i.e., doing robotics in Python. First, thanks a lot for the amazing and impressive interest we got in it. We are now @ 1.13 K downloads, which is beyond my wildest expectations ( https://t.co/eEKILJTK1H)! 0.
1
1
11
You do robotics, you say? ML is the way to go for robotics, you say? ML is there for robotics, it is a matter of years, you say? You say a lot, but can you back it up and prove it? Come show us! We are arranging an amazing competition with @Jan_R_Peters and his team on
1
15
32
🗓️ The IBRL Workshop kicks off tomorrow! 🎉 Join us at @RL_Conference @UAlberta to explore how Inductive Biases can boost 🚀 the performance of RL agents. 📄 Accepted papers: https://t.co/m39oYjEg3Z 📅 Full schedule: https://t.co/2kazxB3692
#ReinforcementLearning #RLC2025
1
13
33
Should we use a target network in deep value-based RL?🤔 The answer has always been YES or NO, as there are pros and cons. @RLFrameWorkshop, I will present iS-QN, a method that lies in between this binary view, collecting the pros while reducing the cons🚀
2
12
102
To increase the reward propagation in value-based RL algorithms, it is tempting to reduce the target update period🤔 But, this makes the training unstable💔 @RL_Conference, I will present i-QN, a new method that allows faster reward propagation, while keeping stability⚡️ 👉🧵
1
12
92
Pushing for #icra but still missing real robot experiments? 😰 Skip the ROS headaches — get your Franka robot running in minutes with franky! 🦾 Super beginner-friendly, Pythonic, and fast to set up. 🔗 https://t.co/Modc3KX2TY
@ias_tudarmstadt @Jan_R_Peters 🧵👇
1
16
67