Search our Publications
Latest Publications
Radu Stoican, Angelo Cangelosi, Christian Goerick, Thomas H Weisswange , "Task-Specific Exploration in Meta-Reinforcement Learning via Task Reconstruction", 5th Conference on Lifelong Learning Agents (CoLLAs 2026): Published Papers Track, 2026.
AbstractReinforcement learning trains policies specialized for a single task. Meta-reinforcement learning (meta-RL) improves upon this by leveraging prior experience to train policies for few-shot adaptation to new tasks. However, existing meta-RL approaches often struggle to explore and learn tasks effectively. We introduce a novel meta-RL algorithm that learns to learn task-specific exploration policies for sample-efficient few-shot adaptation. We achi...
Download Bibtex file
David Rother , "Implicitly Cooperative Agents Through Impact Aware Learning", Technische Universität Darmstadt, Technische Universität Darmstadt, pp. 129, 2026.
AbstractIn recent years, the increasing integration of autonomous agents in human-centric environments has emphasized the importance of advanced interaction methodologies. This dissertation studies the challenges and solutions that arise in the deployment of robots and AI assistants, particularly focusing on the intricacies of human-agent interaction. Despite the advancements in machine learning and neural networks, the scalability and adaptability of ...
Download Bibtex file Per Mail Request
Phillip Richter, Heiko Wersing, Anna-Lisa Vollmer , "Understanding and Addressing Mental Model Mismatches in Human-Robot Teaching", 2nd German Robotics Conference, 2026.
AbstractA major challenge in human-robot interaction is the mental model mismatch, which arises when a human's understanding of a robot's capabilities differs from the robot's actual operational model. Such mismatches can result in ineffective teaching, suboptimal performance, and interaction breakdowns. This work aims to quantify and systematically categorize mental model mismatches by formalizing human expectations and comparing them with robot learnin...
Download Bibtex file Per Mail Request
Phillip Richter, Mara Brandt, Heiko Wersing, Anna-Lisa Vollmer , "Evaluation of Social Robots Blossom and Mirokai Using ASAQ and ASOR", ACM/IEEE International Conference on Human-Robot Interaction Companion, 2026.
AbstractAs social robots enter a variety of human environments, there is a need for systematic evaluation methods to guide deployment decisions. In this study, we evaluated two contrasting social robot platforms using standardised questionnaires: Blossom, an open-source, hand-crafted robot and Mirokai, a commercial, ball-bot humanoid. Participants were randomly assigned to evaluate one of the robots (n = 50 for each robot) after watching video demonstrat...
Download Bibtex file Per Mail Request
Linus Ekstrom, Hao Wang, Sebastian Schmitt , "Comparing Qubit and Qudit Encoding for EV Charging and Trip Assignment Problems", The Genetic and Evolutionary Computation Conference (GECCO) , 2026.
AbstractVariational quantum algorithms have garnered attention for their promise in solving combinatorial optimization problems. We study how the encoding choice impacts resource requirements and optimization behavior of a variational quantum optimization algorithms. We consider a realistic constrained electric vehicle (EV) fleet management problem that couples determining the optimal bidirectional charging schedule with assigning EVs to trips request...
Download Bibtex file Per Mail Request
Laurenz Tomandl, Maria Bresich, Guenther Raidl, Yi Mei, Steffen Limmer, Tobias Rodemann , "Approaching the Dynamic Electric Autonomous Dial-a-Ride Problem with Large Neighborhood Search and Reinforcement Learning", The 24th Conference of the International Federation of Operational Research Societies (IFORS) 2026, 2026.
AbstractThe Dynamic Electric Autonomous Dial-a-Ride Problem addresses shared real-time passenger transport from request-specific origins to destinations. Large Neighborhood Search (LNS) heuristics achieve state-of-the-art results for the static variant of this problem in which all information is given upfront. We extend this framework to the dynamic setting by controlling waiting times and the priorization of charging with a neural network trained by rei...
Download Bibtex file Per Mail Request
Sai Lokesh Kancharla, Sebastian Brulin, Markus Olhofer, Sanaz Mostaghim , "Adaptable Charging Station Placement: Employing Evolvability", IEEE Intelligent Vehicles Symposium, 2026.
AbstractThis work introduces designing adaptable (mobile) Electric Vehicle charging station infrastructure, employing evolvability and dynamic optimization framework to identify layouts that are not only optimal for current traffic and network conditions but also facilitate easy adaptation to future changes. The framework leverages mechanisms of evolvability, including regulatory change, exploration, and diversity, to enhance the adaptability of the solu...
Download Bibtex file
Hao Tong, Leandro L. Minku, Stefan Menzel, Bernhard Sendhoff, Xin Yao , "Building Blocks as Experiences in Dynamic Capacitated Arc Routing Problems ", CAAI Transactions on Intelligence Technology, 2026.
AbstractThe Dynamic Capacitated Arc Routing Problem (DCARP) aims to update the service paths of vehicles in the capacitated arc routing problem when uncertain factors deteriorate the current schedule of vehicles' services. For example, a road may become congested or inaccessible due to a traffic accident, or new tasks may need to be served. A DCARP scenario comprises a series of DCARP instances that share similarities with each other. Therefore, optimiza...
Download Bibtex file
Maria Bresich, Jingyi Peng, Guenther Raidl, Steffen Limmer , "Determining Destroy Sets in Large Neighborhood Search by Generative Flow Networks", Parallel Problem Solving from Nature -- PPSN XIX, pp. 265-281, 2026.
AbstractLarge Neighborhood Search (LNS) is an often applied meta-heuristic to heuristically solve challenging combinatorial optimization problems. Typically, it relies on the scheme of iteratively removing parts of a solution by one or more destroy operators and cleverly augmenting the solution again by repair operators. Destroy operators are usually handcrafted and often a mixture of uniform random selection and problem-specific heuristics, and their ch...
Download Bibtex file Per Mail Request
Johannes Varga, Guenther Raidl, Tobias Rodemann , "Learning to Predict User Replies in Interactive Job Scheduling", Machine Learning, Optimization, and Data Science. LOD 2025. , vol. 16467 , 2026.
AbstractWe consider a learning task that arises within an interactive job scheduling setting, in which a scheduler plans the execution of jobs that require the presence of human users. Availabilities of these users shall be considered but are only known partially, and thus the scheduler presents queries to the users to receive more information about the users' availabilities. Having a precise understanding of typical user behavior is crucial for the eff...
Download Bibtex file Per Mail Request
1 ... 4 5 6 7 8 9 ... 183
Follow us
