Daniel Palenicek

Daniel Palenicek

I am a fifth-year PhD student at TU Darmstadt & hessian.AI supervised by Prof. Jan Peters. During my PhD I have been working on sample-efficient reinforcement learning for robotics. I was selected as a 2026-2027 NVIDIA Graduate Fellowship Finalist.

CV [03/2026] LinkedIn Scholar GitHub X

Short CV

2026—Now Research Scientist Intern Meta FAIR, New York
2021—Now PhD Student TU Darmstadt & hessian.AI
2019—2020 Research Intern Huawei R&D London
2019 Research Intern Bosch Center for AI, Renningen
2015—2021 M.Sc. Wirtschaftsinformatik TU Darmstadt

📄 Selected Publications

Diminishing Return

Diminishing Return of Value Expansion Methods

Palenicek D., Lutter M., Carvalho J., Dennert D., Ahmad F., Peters J.
IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 2026In Press

XQCfD: Accelerating Fast Actor-Critic Algorithms with Prior Data and Prior Policies

Palenicek D.*, Vogt F.*, Watson J.*, Posner I., Kragic D., Peters J.
ICML 2026 Workshop on Decision-Making from Offline Datasets to Online Adaptation: Black-Box Optimization to Reinforcement LearningSpotlight
Paper

XQC: Well-Conditioned Optimization Accelerates Deep Reinforcement Learning

Palenicek D., Vogt F., Watson J., Posner I., Peters J.
ICLR 2026
Paper Project Page

FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control

Kim D.*, Lee Y.*, Park M., Kim K., Nahrendra I.M.A., Seno T., Min S., Palenicek D., Vogt F., Kragic D., Peters J., Choo J., Lee H.
RSS 2026Outstanding Paper Award
Paper Code Project Page

Scaling Off-Policy Reinforcement Learning with Batch and Weight Normalization

Palenicek D., Vogt F., Watson J., Peters J.
NeurIPS 2025
Paper

Gait in Eight: Efficient On-Robot Learning for Omnidirectional Quadruped Locomotion

Bohlinger N.*, Kinzel J.*, Palenicek D., Antczak L., Peters J.
IROS 2025
Paper Project Page

Towards Safe Robot Foundation Models Using Inductive Biases

Palenicek D.*, Tölle M.*, Gruner T.*, Schneider T., Günster J., Watson J., Tateo D., Liu P., Peters J.
Safe-VLM Workshop @ ICRA 2025Spotlight & GRC 2025
Paper Project Page

CrossQ: Batch Normalization in Deep Reinforcement Learning for Greater Sample Efficiency and Simplicity

Palenicek D.*, Bhatt A.*, Belousov B., Argus M., Amiranashvili A., Brox T., Peters J.
ICLR 2024Spotlight
Paper Code Webpage

Learning Tactile Insertion in the Real World

Palenicek D.*, Gruner T.*, Schneider T.*, Böhm A.*, Lenz J.*, Pfenning I.*, Krämer E., Peters J.
ICRA@40 & ViTac Workshop @ ICRA 2024
Paper
Diminishing Return

Diminishing Return of Value Expansion Methods in Model-Based Reinforcement Learning

Palenicek D., Lutter M., Carvalho J., Peters J.
ICLR 2023
Paper Code

See Google Scholar for complete list.

🎤 Talks

Sample Efficiency in Deep RL: Quo Vadis? / link

BeNeRL Seminar Series, September 2024

CrossQ: Batch Normalization in Deep Reinforcement Learning / link

ML&AI Academy, February 2024

AI and its faces (interview) / link

hessian.AI, with Theo Gruner, August 2023

Slurm-tastic Adventures on the IAS Cluster / link

IWIALS (Workshop on Intelligent Autonomous Systems), August 2023