Publications (126)
MP3: Movement Primitive-Based (Re-)Planning Policy
Fabian Otto, Hongyi Zhou, Onur Celik +3
We introduce a novel deep reinforcement learning (RL) approach called Movement Primitive-based Planning Policy (MP3). By integrating movement primitives (MPs) into the deep RL fram…
Grounding Graph Network Simulators using Physical Sensor Observations
Jonas Linkerhägner, Niklas Freymuth, Paul Maria Scheikl +2
Physical simulations that accurately model reality are crucial for many engineering disciplines such as mechanical engineering and robotic motion planning. In recent years, learned…
VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models
Florian Seligmann, Emiliyan Gospodinov, Enes Ulas Dincer +1
Vision-language-action models (VLAs) achieve state-of-the-art performance on many robotic manipulation tasks, yet they can still behave unpredictably in out-of-distribution scenari…
Combining Reconstruction and Contrastive Methods for Multimodal Representations in RL
Philipp Becker, Sebastian Mossburger, Fabian Otto +1
Learning self-supervised representations using reconstruction or contrastive losses improves performance and sample complexity of image-based and multimodal reinforcement learning…
Switching Recurrent Kalman Networks
Giao Nguyen-Quynh, Philipp Becker, Chen Qiu +2
Forecasting driving behavior or other sensor measurements is an essential component of autonomous driving systems. Often real-world multivariate time series data is hard to model b…
Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarios
Vaisakh Shaj, Dieter Buchler, Rohit Sonker +2
Recurrent State-space models (RSSMs) are highly expressive models for learning patterns in time series data and system identification. However, these models assume that the dynamic…
Imitation Learning for Autonomous Trajectory Learning of Robot Arms in Space
RB Ashith Shyam, Zhou Hao, Umberto Montanaro +1
This work adds on to the on-going efforts to provide more autonomy to space robots. Here the concept of programming by demonstration or imitation learning is used for trajectory pl…
Beyond Task Performance: Human Experience in Human-Robot Collaboration
Sean Kille, Jan Heinrich Robens, Philipp Dahlinger +8
Human interaction experience plays a crucial role in the effectiveness of human-machine collaboration, especially as interactions in future systems progress towards tighter physica…
Residual Feedback Learning for Contact-Rich Manipulation Tasks with Uncertainty
Alireza Ranjbar, Ngo Anh Vien, Hanna Ziesche +2
While classic control theory offers state of the art solutions in many problem scenarios, it is often desired to improve beyond the structure of such solutions and surpass their li…
PointPatchRL -- Masked Reconstruction Improves Reinforcement Learning on Point Clouds
Balázs Gyenes, Nikolai Franke, Philipp Becker +1
Perceiving the environment via cameras is crucial for Reinforcement Learning (RL) in robotics. While images are a convenient form of representation, they often complicate extractin…
Navigate-and-Seek: a Robotics Framework for People Localization in Agricultural Environments
Riccardo Polvara, Francesco Del Duchetto, Gerhard Neumann +1
The agricultural domain offers a working environment where many human laborers are nowadays employed to maintain or harvest crops, with huge potential for productivity gains throug…
MuTT: A Multimodal Trajectory Transformer for Robot Skills
Claudius Kienle, Benjamin Alt, Onur Celik +4
High-level robot skills represent an increasingly popular paradigm in robot programming. However, configuring the skills' parameters for a specific task remains a manual and time-c…
PAWS: Preference Learning with Advantage-Weighted Segments
Aleksandar Taranovic, Onur Celik, Niklas Freymuth +6
Preference-based reinforcement learning (PbRL) learns policies from human trajectory-level comparisons, avoiding explicit reward design and expert demonstrations. Existing methods…
BMP: Bridging the Gap between B-Spline and Movement Primitives
Weiran Liao, Ge Li, Hongyi Zhou +2
This work introduces B-spline Movement Primitives (BMPs), a new Movement Primitive (MP) variant that leverages B-splines for motion representation. B-splines are a well-known conce…
Specializing Versatile Skill Libraries using Local Mixture of Experts
Onur Celik, Dongzhuoran Zhou, Ge Li +2
A long-cherished vision in robotics is to equip robots with skills that match the versatility and precision of humans. For example, when playing table tennis, a robot should be cap…
Beyond ELBOs: A Large-Scale Evaluation of Variational Methods for Sampling
Denis Blessing, Xiaogang Jia, Johannes Esslinger +2
Monte Carlo methods, Variational Inference, and their combinations play a pivotal role in sampling from intractable probability distributions. However, current studies lack a unifi…
Trust-Region Diffusion Policies for Massively Parallel On-Policy RL
Huy Le, Onur Celik, Denis Blessing +6
Reinforcement learning with massively parallel simulations has become a standard framework for developing robust, deployable policies; however, most existing approaches still rely…
Reactive Motion Generation on Learned Riemannian Manifolds
Hadi Beik-Mohammadi, Søren Hauberg, Georgios Arvanitidis +2
In recent decades, advancements in motion learning have enabled robots to acquire new skills and adapt to unseen conditions in both structured and unstructured environments. In pra…
MaNGO - Adaptable Graph Network Simulators via Meta-Learning
Philipp Dahlinger, Tai Hoang, Denis Blessing +2
Accurately simulating physics is crucial across scientific domains, with applications spanning from robotics to materials science. While traditional mesh-based simulations are prec…
Variational Distillation of Diffusion Policies into Mixture of Experts
Hongyi Zhou, Denis Blessing, Ge Li +4
This work introduces Variational Diffusion Distillation (VDD), a novel method that distills denoising diffusion policies into Mixtures of Experts (MoE) through variational inferenc…
AMBER: Adaptive Mesh Generation by Iterative Mesh Resolution Prediction
Niklas Freymuth, Tobias Würth, Nicolas Schreiber +9
The cost and accuracy of simulating complex physical systems using the Finite Element Method (FEM) scales with the resolution of the underlying mesh. Adaptive meshes improve comput…
TROLL: Trust Regions improve Reinforcement Learning for Large Language Models
Philipp Becker, Niklas Freymuth, Serge Thilges +2
Reinforcement Learning (RL) with PPO-like clip objectives has become the standard choice for reward-based fine-tuning of large language models (LLMs). Although recent work has expl…
Differentiable Trust Region Layers for Deep Reinforcement Learning
Fabian Otto, Philipp Becker, Ngo Anh Vien +2
Trust region methods are a popular tool in reinforcement learning as they yield robust policy updates in continuous and discrete action spaces. However, enforcing such trust region…
Guided Deep Reinforcement Learning for Swarm Systems
Maximilian Hüttenrauch, Adrian Å oÅ¡iÄ, Gerhard Neumann
In this paper, we investigate how to learn to control a group of cooperative agents with limited sensing capabilities such as robot swarms. The agents have only very basic sensor c…
DMFC-GraspNet: Differentiable Multi-Fingered Robotic Grasp Generation in Cluttered Scenes
Philipp Blättner, Johannes Brand, Gerhard Neumann +1
Robotic grasping is a fundamental skill required for object manipulation in robotics. Multi-fingered robotic hands, which mimic the structure of the human hand, can potentially per…
Recurrent Kalman Networks: Factorized Inference in High-Dimensional Deep Feature Spaces
Philipp Becker, Harit Pandya, Gregor Gebhardt +3
In order to integrate uncertainty estimates into deep time-series modelling, Kalman Filters (KFs) (Kalman et al., 1960) have been integrated with deep learning models, however, suc…
Swarm Reinforcement Learning For Adaptive Mesh Refinement
Niklas Freymuth, Philipp Dahlinger, Tobias Würth +3
Adaptive Mesh Refinement (AMR) enhances the Finite Element Method, an important technique for simulating complex problems in engineering, by dynamically refining mesh regions, enab…
TOP-ERL: Transformer-based Off-Policy Episodic Reinforcement Learning
Ge Li, Dong Tian, Hongyi Zhou +3
This work introduces Transformer-based Off-Policy Episodic Reinforcement Learning (TOP-ERL), a novel algorithm that enables off-policy updates in the ERL framework. In ERL, policie…
Non-Adversarial Imitation Learning and its Connections to Adversarial Methods
Oleg Arenz, Gerhard Neumann
Many modern methods for imitation learning and inverse reinforcement learning, such as GAIL or AIRL, are based on an adversarial formulation. These methods apply GANs to match the…
Towards Near-Real-Time Telemetry-Aware Routing with Neural Routing Algorithms
Andreas Boltres, Niklas Freymuth, Benjamin Schichtholz +2
Routing algorithms are crucial for efficient computer network operations, and in many settings they must be able to react to traffic bursts within milliseconds. Live telemetry data…
Curriculum-Based Imitation of Versatile Skills
Maximilian Xiling Li, Onur Celik, Philipp Becker +3
Learning skills by imitation is a promising concept for the intuitive teaching of robots. A common way to learn such skills is to learn a parametric model by maximizing the likelih…
Learning Sub-Second Routing Optimization in Computer Networks requires Packet-Level Dynamics
Andreas Boltres, Niklas Freymuth, Patrick Jahnke +2
Finding efficient routes for data packets is an essential task in computer networking. The optimal routes depend greatly on the current network topology, state and traffic demand,…
PointMapPolicy: Structured Point Cloud Processing for Multi-Modal Imitation Learning
Xiaogang Jia, Qian Wang, Anrui Wang +12
Robotic manipulation systems benefit from complementary sensing modalities, where each provides unique environmental information. Point clouds capture detailed geometric structure,…
Action-Conditional Recurrent Kalman Networks For Forward and Inverse Dynamics Learning
Vaisakh Shaj, Philipp Becker, Dieter Buchler +5
Estimating accurate forward and inverse dynamics models is a crucial component of model-based control for sophisticated robots such as robots driven by hydraulics, artificial muscl…
Adaptive Swarm Mesh Refinement using Deep Reinforcement Learning with Local Rewards
Niklas Freymuth, Philipp Dahlinger, Tobias Würth +3
Simulating physical systems is essential in engineering, but analytical solutions are limited to straightforward problems. Consequently, numerical methods like the Finite Element M…
Underdamped Diffusion Bridges with Applications to Sampling
Denis Blessing, Julius Berner, Lorenz Richter +1
We provide a general framework for learning diffusion bridges that transport prior to target distributions. It includes existing diffusion models for generative modeling, but also…
Physics-informed MeshGraphNets (PI-MGNs): Neural finite element solvers for non-stationary and nonlinear simulations on arbitrary meshes
Tobias Würth, Niklas Freymuth, Clemens Zimmerling +2
Engineering components must meet increasing technological demands in ever shorter development cycles. To face these challenges, a holistic approach is essential that allows for the…
Model-Free Trajectory-based Policy Optimization with Monotonic Improvement
Riad Akrour, Abbas Abdolmaleki, Hany Abdulsamad +2
Many of the recent trajectory optimization algorithms alternate between linear approximation of the system dynamics around the mean trajectory and conservative policy update. One w…
Diffusion-Based Hierarchical Graph Neural Networks for Simulating Nonlinear Solid Mechanics
Tobias Würth, Niklas Freymuth, Gerhard Neumann +1
Graph-based learned simulators have emerged as a promising approach for simulating physical systems on unstructured meshes, offering speed and generalization across diverse geometr…
MaIL: Improving Imitation Learning with Mamba
Xiaogang Jia, Qian Wang, Atalay Donat +7
This work presents Mamba Imitation Learning (MaIL), a novel imitation learning (IL) architecture that provides an alternative to state-of-the-art (SoTA) Transformer-based policies.…
Neural Contractive Dynamical Systems
Hadi Beik-Mohammadi, Søren Hauberg, Georgios Arvanitidis +3
Stability guarantees are crucial when ensuring a fully autonomous robot does not take undesirable or potentially harmful actions. Unfortunately, global stability guarantees are har…
Probabilistic approach to physical object disentangling
Joni Pajarinen, Oleg Arenz, Jan Peters +1
Physically disentangling entangled objects from each other is a problem encountered in waste segregation or in any task that requires disassembly of structures. Often there are no…
DIME:Diffusion-Based Maximum Entropy Reinforcement Learning
Onur Celik, Zechu Li, Denis Blessing +5
Maximum entropy reinforcement learning (MaxEnt-RL) has become the standard approach to RL due to its beneficial exploration properties. Traditionally, policies are parameterized us…
What Matters For Meta-Learning Vision Regression Tasks?
Ning Gao, Hanna Ziesche, Ngo Anh Vien +2
Meta-learning is widely used in few-shot classification and function regression due to its ability to quickly adapt to unseen tasks. However, it has not yet been well explored on r…
Robot-DIFT: Correspondence-Sensitive Diffusion Features for Contact-Rich Robot Manipulation
Yu Deng, Yufeng Jin, Xiaogang Jia +3
Robot manipulation often fails in the final millimeters: a policy may recognize the right object yet miss the pose offsets, boundaries, or pre-contact alignments needed for action.…
Beyond Deep Ensembles: A Large-Scale Evaluation of Bayesian Deep Learning under Distribution Shift
Florian Seligmann, Philipp Becker, Michael Volpp +1
Bayesian deep learning (BDL) is a promising approach to achieve well-calibrated predictions on distribution-shifted data. Nevertheless, there exists no large-scale survey that eval…
Hierarchical Policy Learning for Mechanical Search
Oussama Zenkri, Ngo Anh Vien, Gerhard Neumann
Retrieving objects from clutters is a complex task, which requires multiple interactions with the environment until the target object can be extracted. These interactions involve e…
Hybrid control trajectory optimization under uncertainty
Joni Pajarinen, Ville Kyrki, Michael Koval +3
Trajectory optimization is a fundamental problem in robotics. While optimization of continuous control trajectories is well developed, many applications require both discrete and c…
Nautilus: From One Prompt to Plug-and-Play Robot Learning
Yufeng Jin, Jianfei Guo, Xiaogang Jia +8
Robot learning research is fragmented across policy families, benchmark suites, and real robots; each implementation is entangled with the others in a complex combination matrix, m…
MV6D: Multi-View 6D Pose Estimation on RGB-D Frames Using a Deep Point-wise Voting Network
Fabian Duffhauss, Tobias Demmler, Gerhard Neumann
Estimating 6D poses of objects is an essential computer vision task. However, most conventional approaches rely on camera data from a single perspective and therefore suffer from o…
Sequential Controlled Langevin Diffusions
Junhua Chen, Lorenz Richter, Julius Berner +3
An effective approach for sampling from unnormalized densities is based on the idea of gradually transporting samples from an easy prior to the complicated target distribution. Two…
Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations
Xiaogang Jia, Denis Blessing, Xinkai Jiang +4
Imitation learning with human data has demonstrated remarkable success in teaching robots in a wide range of skills. However, the inherent diversity in human behavior leads to the…
SEAR: Sample Efficient Action Chunking Reinforcement Learning
C. F. Maximilian Nagy, Onur Celik, Emiliyan Gospodinov +4
Action chunking can improve exploration and value estimation in long horizon reinforcement learning, but makes learning substantially harder since the critic must evaluate action s…
An Algorithmic Perspective on Imitation Learning
Takayuki Osa, Joni Pajarinen, Gerhard Neumann +3
As robots and other intelligent agents move from simple environments and problems to more complex, unstructured settings, manually programming their behavior has become increasingl…
End-to-End Learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control
Moritz Reuss, Niels van Duijkeren, Robert Krug +3
It is well-known that inverse dynamics models can improve tracking performance in robot control. These models need to precisely capture the robot dynamics, which consist of well-un…
Efficient Off-Policy Learning for High-Dimensional Action Spaces
Fabian Otto, Philipp Becker, Ngo Anh Vien +1
Existing off-policy reinforcement learning algorithms often rely on an explicit state-action-value function representation, which can be problematic in high-dimensional action spac…
Agentic Language-to-Objective Synthesis for Optofluidic Assembly
Ivan Saraev, Elena Erben, Weida Liao +4
Light-based advanced manufacturing increasingly requires programmable, closed-loop tools that translate human design intent into executable operations at small length scales. Yet a…
Category-Agnostic 6D Pose Estimation with Conditional Neural Processes
Yumeng Li, Ning Gao, Hanna Ziesche +1
We present a novel meta-learning approach for 6D pose estimation on unknown objects. In contrast to ``instance-level" and ``category-level" pose estimation methods, our algorithm l…
Fourier Features Let Agents Learn High Precision Policies with Imitation Learning
Balázs Gyenes, Emiliyan Gospodinov, Jan Frieling +5
High-precision robotic manipulation requires fine-grained spatial reasoning that is often difficult to achieve with RGB-only policies due to depth ambiguity and perspective scale i…
Bridge Matching Sampler: Scalable Sampling via Generalized Fixed-Point Diffusion Matching
Denis Blessing, Lorenz Richter, Julius Berner +2
Sampling from unnormalized densities using diffusion models has emerged as a powerful paradigm. However, while recent approaches that use least-squares `matching' objectives have i…
SyMFM6D: Symmetry-aware Multi-directional Fusion for Multi-View 6D Object Pose Estimation
Fabian Duffhauss, Sebastian Koch, Hanna Ziesche +2
Detecting objects and estimating their 6D poses is essential for automated systems to interact safely with the environment. Most 6D pose estimators, however, rely on a single camer…
RayViT: Ray-Conditioned Visual Representations for Viewpoint-Robust Imitation Learning
Qian Wang, Longrui Chen, Peiran Sun +8
Visual imitation learning enables robots to acquire visuomotor skills directly from images, yet RGB observations lack explicit geometric cues, making learned policies brittle to ca…
A Multi-View 3D Telepresence System for XR Robot Teleoperation
Enes Ulas Dincer, Manuel Zaremski, Alexandra Nick +3
Robot teleoperation is critical for applications such as remote maintenance, fleet robotics, search and rescue, and data collection for robot learning. Effective teleoperation requ…
Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors
Niklas Freymuth, Nicolas Schreiber, Philipp Becker +2
Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt…
Compatible Natural Gradient Policy Search
Joni Pajarinen, Hong Linh Thai, Riad Akrour +2
Trust-region methods have yielded state-of-the-art results in policy search. A common approach is to use KL-divergence to bound the region of trust resulting in a natural gradient…
Differentiable Robust LQR Layers
Ngo Anh Vien, Gerhard Neumann
This paper proposes a differentiable robust LQR layer for reinforcement learning and imitation learning under model uncertainty and stochastic dynamics. The robust LQR layer can ex…
Scaffolding Dexterous Manipulation with Vision-Language Models
Vincent de Bakker, Joey Hejna, Tyler Ga Wei Lum +6
Dexterous robotic hands are essential for performing complex manipulation tasks, yet remain difficult to train due to the challenges of demonstration collection and high-dimensiona…
Towards Fusing Point Cloud and Visual Representations for Imitation Learning
Atalay Donat, Xiaogang Jia, Xi Huang +7
Learning for manipulation requires using policies that have access to rich sensory information such as point clouds or RGB images. Point clouds efficiently capture geometric struct…
Trust Region Constrained Measure Transport in Path Space for Stochastic Optimal Control and Inference
Denis Blessing, Julius Berner, Lorenz Richter +4
Solving stochastic optimal control problems with quadratic control costs can be viewed as approximating a target path space measure, e.g. via gradient-based optimization. In practi…
Adaptation and Robust Learning of Probabilistic Movement Primitives
Sebastian Gomez-Gonzalez, Gerhard Neumann, Bernhard Schölkopf +1
Probabilistic representations of movement primitives open important new possibilities for machine learning in robotics. These representations are able to capture the variability of…
Smooth Sampling-Based Model Predictive Control Using Deterministic Samples
Markus Walker, Marcel Reith-Braun, Tai Hoang +2
Sampling-based model predictive control (MPC) is effective for nonlinear systems but often produces non-smooth control inputs due to random sampling. To address this issue, we exte…
Meta-Learning Regrasping Strategies for Physical-Agnostic Objects
Ning Gao, Jingyu Zhang, Ruijie Chen +3
Grasping inhomogeneous objects in real-world applications remains a challenging task due to the unknown physical properties such as mass distribution and coefficient of friction. I…
Learning Riemannian Manifolds for Geodesic Motion Skills
Hadi Beik-Mohammadi, Søren Hauberg, Georgios Arvanitidis +2
For robots to work alongside humans and perform in unstructured environments, they must learn new motion skills and adapt them to unseen situations on the fly. This demands learnin…
Geometry-aware RL for Manipulation of Varying Shapes and Deformable Objects
Tai Hoang, Huy Le, Philipp Becker +2
Manipulating objects with varying geometries and deformable objects is a major challenge in robotics. Tasks such as insertion with different objects or cloth hanging require precis…
End-To-End Learning of Gaussian Mixture Priors for Diffusion Sampler
Denis Blessing, Xiaogang Jia, Gerhard Neumann
Diffusion models optimized via variational inference (VI) have emerged as a promising tool for generating samples from unnormalized target densities. These models create samples by…
Deep Black-Box Reinforcement Learning with Movement Primitives
Fabian Otto, Onur Celik, Hongyi Zhou +3
\Episode-based reinforcement learning (ERL) algorithms treat reinforcement learning (RL) as a black-box optimization problem where we learn to select a parameter vector of a contro…
A Retrospective on the Robot Air Hockey Challenge: Benchmarking Robust, Reliable, and Safe Learning Techniques for Real-world Robotics
Puze Liu, Jonas Günster, Niklas Funk +17
Machine learning methods have a groundbreaking impact in many application domains, but their application on real robotic platforms is still limited. Despite the many challenges ass…
Iterative Sizing Field Prediction for Adaptive Mesh Generation From Expert Demonstrations
Niklas Freymuth, Philipp Dahlinger, Tobias Würth +5
Many engineering systems require accurate simulations of complex physical systems. Yet, analytical solutions are only available for simple problems, necessitating numerical approxi…
Local Communication Protocols for Learning Complex Swarm Behaviors with Deep Reinforcement Learning
Maximilian Hüttenrauch, Adrian Å oÅ¡iÄ, Gerhard Neumann
Swarm systems constitute a challenging problem for reinforcement learning (RL) as the algorithm needs to learn decentralized control policies that can cope with limited local sensi…
KalMamba: Towards Efficient Probabilistic State Space Models for RL under Uncertainty
Philipp Becker, Niklas Freymuth, Gerhard Neumann
Probabilistic State Space Models (SSMs) are essential for Reinforcement Learning (RL) from high-dimensional, partial information as they provide concise representations for control…
Movement Primitive Diffusion: Learning Gentle Robotic Manipulation of Deformable Objects
Paul Maria Scheikl, Nicolas Schreiber, Christoph Haas +4
Policy learning in robot-assisted surgery (RAS) lacks data efficient and versatile methods that exhibit the desired motion quality for delicate surgical interventions. To this end,…
ATLAS: A Foundation Neural Sampler for Amorphous Materials
Mouyang Cheng, Denis Blessing, Botao Yu +4
Amorphous materials exhibit exceptional mechanical and functional properties, yet their rugged energy landscapes are notoriously difficult to sample. Below the glass-transition tem…
On Uncertainty in Deep State Space Models for Model-Based Reinforcement Learning
Philipp Becker, Gerhard Neumann
Improved state space models, such as Recurrent State Space Models (RSSMs), are a key factor behind recent advances in model-based reinforcement learning (RL). Yet, despite their em…
Regret-Aware Black-Box Optimization with Natural Gradients, Trust-Regions and Entropy Control
Maximilian Hüttenrauch, Gerhard Neumann
Most successful stochastic black-box optimizers, such as CMA-ES, use rankings of the individual samples to obtain a new search distribution. Yet, the use of rankings also introduce…
Learning Boltzmann Generators via Constrained Mass Transport
Christopher von Klitzing, Denis Blessing, Henrik Schopmans +2
Efficient sampling from high-dimensional and multimodal unnormalized probability distributions is a central challenge in many areas of science and machine learning. We focus on Bol…
Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Experts
Onur Celik, Aleksandar Taranovic, Gerhard Neumann
Reinforcement learning (RL) is a powerful approach for acquiring a good-performing policy. However, learning diverse skills is challenging in RL due to the commonly used Gaussian p…
Versatile Inverse Reinforcement Learning via Cumulative Rewards
Niklas Freymuth, Philipp Becker, Gerhard Neumann
Inverse Reinforcement Learning infers a reward function from expert demonstrations, aiming to encode the behavior and intentions of the expert. Current approaches usually do this w…
Point Cloud Segmentation for Autonomous Clip Positioning in Laparoscopic Cholecystectomy on a Phantom
Balázs Gyenes, Nikolai Franke, Paul Maria Scheikl +5
High-risk applications in robotics, such as robot-assisted surgery, present unique challenges. These systems must be both highly precise and interpretable in order to be deployed i…
Adaptive World Models: Learning Behaviors by Latent Imagination Under Non-Stationarity
Emiliyan Gospodinov, Vaisakh Shaj, Philipp Becker +2
Developing foundational world models is a key research direction for embodied intelligence, with the ability to adapt to non-stationary environments being a crucial criterion. In t…
Chunking the Critic: A Transformer-based Soft Actor-Critic with N-Step Returns
Dong Tian, Onur Celik, Gerhard Neumann
We introduce a sequence-conditioned critic for Soft Actor-Critic (SAC) that models trajectory context with a lightweight Transformer and trains on aggregated -step targets. Unli…
Registered and Segmented Deformable Object Reconstruction from a Single View Point Cloud
Pit Henrich, Balázs Gyenes, Paul Maria Scheikl +2
In deformable object manipulation, we often want to interact with specific segments of an object that are only defined in non-deformed models of the object. We thus require a syste…
A Unified Perspective on Natural Gradient Variational Inference with Gaussian Mixture Models
Oleg Arenz, Philipp Dahlinger, Zihan Ye +2
Variational inference with Gaussian mixture models (GMMs) enables learning of highly tractable yet multi-modal approximations of intractable target distributions with up to a few h…
Towards a Multi-Embodied Grasping Agent
Roman Freiberg, Alexander Qualmann, Ngo Anh Vien +1
Multi-embodiment grasping focuses on developing approaches that exhibit generalist behavior across diverse gripper designs. Existing methods often learn the kinematic structure of…
Robot Policy Learning from Demonstration Using Advantage Weighting and Early Termination
Abdalkarim Mohtasib, Gerhard Neumann, Heriberto Cuayahuitl
Learning robotic tasks in the real world is still highly challenging and effective practical solutions remain to be found. Traditional methods used in this area are imitation learn…
Point Cloud Sequence Encoding for Material-conditioned Graph Network Simulators
Philipp Dahlinger, Balázs Gyenes, Niklas Freymuth +6
Graph Network Simulators (GNSs) have emerged as powerful surrogates for complex physics-based simulation, offering inherent differentiability and orders-of-magnitude speedups over…
Expected Information Maximization: Using the I-Projection for Mixture Density Estimation
Philipp Becker, Oleg Arenz, Gerhard Neumann
Modelling highly multi-modal data is a challenging problem in machine learning. Most algorithms are based on maximizing the likelihood, which corresponds to the M(oment)-projection…
Building Deep Graph Predictors with Graph Imitation Learning
André Eberhard, Gerhard Neumann, Pascal Friederich
Recent years have seen substantial progress in neural generation of text, images, and audio, supported by mature training pipelines and large-scale optimization. For graphs, howeve…
Policy Search with High-Dimensional Context Variables
Voot Tangkaratt, Herke van Hoof, Simone Parisi +3
Direct contextual policy search methods learn to improve policy parameters and simultaneously generalize these parameters to different context or task variables. However, learning…
Domain-Specific Fine-Tuning of Large Language Models for Interactive Robot Programming
Benjamin Alt, Urs KeÃner, Aleksandar Taranovic +4
Industrial robots are applied in a widening range of industries, but robot programming mostly remains a task limited to programming experts. We propose a natural language-based ass…
Latent Task-Specific Graph Network Simulators
Philipp Dahlinger, Niklas Freymuth, Michael Volpp +2
Simulating dynamic physical interactions is a critical challenge across multiple scientific domains, with applications ranging from robotics to material science. For mesh-based sim…