Last week, we had our first lab retreat since the pandemic, enjoying a week in the Austrian mountains and Jan Peters 's birthday! ⛰️🤖⛰️

Few days left to sign up for the new courses Computer Science, TU Darmstadt. Excited to offer 20-00-1158-se and 20-00-1159-pr a Seminar and Practical Lab on Intelligent Robot Manipulation. Sign up if you want to learn more about robot perception, planning and control! TU Darmstadt #Robotics

I am heading to Japan to represent Computer Science, TU Darmstadt TU Darmstadt hessian.AI at #IROS2022 in Kyoto, together with an amazing team from the iROSA and Intelligent Autonomous Systems Group groups! For me this is the most exciting conference so far! See in the following thread why👇 (1/5)👇

The Aalto Robot Learning research group is seeking a talented #postdoc for developing Reinforcement Learning and Planning methods for hw accelerator design. Apply here: aalto.fi/en/open-positi… #Reinforcementlearning #MachineLearning #research #Robotics

We are searching for excellent ML+Robotics PhD students for five different projects at Intelligent Autonomous Systems Group, Computer Science, TU Darmstadt, TU Darmstadt, DFKI, hessian.AI !!! Apply using the link below: ias.informatik.tu-darmstadt.de/Jobs/OpenPosit… Enjoy #phdlife and Research in Germany (Research in Germany - Initiative of the BMBF).

The KI Newcomer*innen 2023 competition of Gesellschaft für Informatik e.V. started! Please vote for the outstanding candidates of Computer Science, TU Darmstadt, TU Darmstadt, hessian.AI, DFKI ... especially: (1/4)

Open positions alert! The iROSA group Computer Science, TU Darmstadt hessian.AI is currently looking for motivated researchers, interested in Embodied AI #Robotics! Follow the link for more information. Please retweet :) #PhDPosition #AcademicJobs irosalab.com/2023/02/08/ope…

Congratulations, 🍾 Roberto! PhD student 007 of Intelligent Autonomous Systems Group !

Congratulations 🎉🎈🎊🍾 , Prof. Carlo D'Eramo! You are the 18. Professor originating from Intelligent Autonomous Systems Group !

After receiving 4 offers, I am excited to announce that I have decided to keep pursuing my research goals at TU Darmstadt as Full Professor sponsored by hessian.AI. I am grateful to my mentors, family, and group for their support. #AI #Robotics (1/2)

#UAI2024 Interested in how power mean can enhance value function estimation in tree search methods? Learn about our approach to solving biases in MCTS for stochastic settings. Join us tomorrow at 4:30 PM in the Exhibition room, building 20. Paper: arxiv.org/pdf/2406.02235

#ICML2024 Come check out these posters at the @FoRLac_workshop: Power Mean Estimation to alleviate bias in Stochastic MCTS: openreview.net/pdf?id=jdCChyt… Monte-Carlo Planning with Thompson Sampling to improve simple regret in Stochastic MCTS: openreview.net/pdf?id=HRbjVO1…

Good Job, Tuan Dam!

Excited to share our work accepted as #ICML2025 poster! Ever wondered how AlphaGo-like algorithms can be used beyond games? Good news! We've adapted #MCTS, a powerful planner, for environments with ambiguities in dynamics and rewards. paper: openreview.net/forum?id=m25ma… #RL #AI 1/2

This work was born out of #icml2024; where I met Tuan Dam! happy to see this appear at #icml2025! Work with Tuan, Brahim Driss, Adam Wierman I am not attending (read: recuperating from a hard job search season). Meet Tuan on 15 July at 4.30pm. Please visit the poster! 2/2

🎉 A new paper for #ICML2025: Power-Mean MCTS → first guarantees for continuous stochastic MDPs 📷• Power-mean backup (p-tunable) • Poly. exploration, O(n^{-\zeta}) conver. • 3× wins on robotics tasks 📷#W817 ▸ Wed 16 Jul 4 : 30 pm PDT #RL #MCTS #ContinuousControl

🎉 New #ICML2025 paper! Online Robust RL via Monte-Carlo Planning — first MCTS that plans for worst-case transitions and rewards, yet keeps the classic O(n^{-1/2}) rate. Work with @kishanpanaganti Adam Wierman & Brahim Driss. 📍 Poster #W720 • 15 Jul 4:30 pm PDT #MCTS #RobustRL

🚀 New paper at #ICML2025! Wasserstein MCTS propagates full uncertainty via L^{1}-Wasserstein barycenters 📷 better planning in stochastic/POMDP settings + O(n^{-1/2})convergence. Meet Tuan Dam on 15 Jul 4 : 30 pm — poster session West Exhi.Hall B2-B3 #W-718. #RL #MCTS #OT