21 papers
The Unreasonable Effectiveness of Discrete-Time Gaussian Process Mixtures for Robot Policy Learning
Jan Ole von Hartz, Adrian Röfer, Joschka Boedecker +1
We present Mixture of Discrete-time Gaussian Processes (MiDiGap), a novel approach for flexible policy representation and imitation learning in robot manipulation. MiDiGap enables…
Probabilistic Recurrent Intention Switching Model
Wenyuan Sheng, Hao Zhu, Joschka Boedecker
Inverse reinforcement learning (IRL) recovers reward functions from observed behavior, yet traditional methods assume a single stationary reward that cannot capture goal switching…
Beyond Self-Play and Scale: A Behavior Benchmark for Generalization in Autonomous Driving
Aron Distelzweig, Faris Janjoš, Andreas Look +7
Recent Autonomous Driving (AD) works such as GigaFlow and PufferDrive have unlocked Reinforcement Learning (RL) at scale as a training strategy for driving policies. Yet such polic…
Spectral Alignment in Forward-Backward Representations via Temporal Abstraction
Seyed Mahdi B. Azad, Jasper Hoffmann, Iman Nematollahi +3
Forward-backward (FB) representations provide a powerful framework for learning the successor representation (SR) in continuous spaces by enforcing a low-rank factorization. Howeve…
Latent Linear Quadratic Regulator for Robotic Control Tasks
Yuan Zhang, Shaohui Yang, Toshiyuki Ohtsuka +2
Model predictive control (MPC) has played a more crucial role in various robotic control tasks, but its high computational requirements are concerning, especially for nonlinear dyn…
MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation
Jan Ole von Hartz, Lukas Schweizer, Joschka Boedecker +1
Generative robot policies such as Flow Matching offer flexible, multi-modal policy learning but are sample-inefficient. Although object-centric policies improve sample efficiency,…