David Rother

David joined the Intelligent Autonomous System lab on October, 1st, 2021 as a PhD student. He is working in a joint project with and at the Honda Research Institute Europe (HRI) in Offenbach am Main, supervised by Jan Peters and Thomas Weisswange on implicitly cooperative robots. He is interested in making robots able to address situations with non-user humans, which require complex decision making.

Before joining the Autonomous Systems Labs, David Rother completed his Bachelor's and Master's degree in Computer Science at the Technische Universitaet Darmstadt. His thesis "Reinforcement Learning in Decentralized Multi-Goal Multi-Agent Settings." was written under supervision of Fabio Muratore and Jan Peters. # (:titlesearch other)

David's research focuses on decision-making for interactive AI agents in multi-agent systems. To that end, he focuses on aspects of optimal decision making including theory of mind models to take humans' values, intentions, and desires into account. His algorithms focus on mixed motive scenarios where any number of other agents may be present, each pursuing their own task. By scaling reinforcement learning through the decomposition of the learning process David hopes to make reinforcement learning applicable in exciting new complex multi-agent situations, which have been prohibitively costly to train and/or perform inference in.

Research Interest

Robotics, Reinforcement Learning, Multi-Agent Learning, Intrinsic Motivation, Machine Learning


    •     Bib
      Ewerton, M.; Rother, D.; Weimar, J.; Kollegger, G.; Wiemeyer, J.; Peters, J.; Maeda, G. (2018). Assisting Movement Training and Execution with Visual and Haptic Feedback, Frontiers in Neurorobotics.
    •     Bib
      Rother, D., Haider, T., & Eger, S (2020). CMCE at SemEval-2020 Task 1: Clustering on Manifolds of Contextualized Embeddings to Detect Historical Meaning Shifts, 14th International Workshop on Semantic Evaluation (SemEval), pp.187-193.
    •     Bib
      Rother, D.; Weisswange, T.H.; Peters, J. (2023). Disentangling Interaction using Maximum Entropy Reinforcement Learning in Multi-Agent Systems, European Conference on Artificial Intelligence (ECAI).
    •     Bib
      Rother, D. (2023). Implicitly Cooperative Agents through Impact-Aware Learning, European Conference on Artificial Intelligence (ECAI).