Dipl.-Inf.

Roscher, Karsten

Filters

4
3 1
Reset filters

Settings
Now showing 1 - 4 of 4
  • Publication
    Domain Shifts in Reinforcement Learning: Identifying Disturbances in Environments
    ( 2021)
    Haider, Tom
    ;
    Schmoeller Roza, Felippe
    ;
    Eilers, Dirk
    ;
    Roscher, Karsten
    ;
    Günnemann, Stephan
    End-to-End Deep Reinforcement Learning (RL) systems return an action no matter what situation they are confronted with, even for situations that differ entirely from those an agent has been trained for. In this work, we propose to formalize the changes in the environment in terms of the Markov Decision Process (MDP), resulting in a more formal framework when dealing with such problems.
  • Publication
    Domain Shifts in Reinforcement Learning: Identifying Disturbances in Environments
    ( 2021)
    Haider, Tom
    ;
    Schmoeller Roza, Felippe
    ;
    Eilers, Dirk
    ;
    Roscher, Karsten
    ;
    Günnemann, Stephan
    A significant drawback of End-to-End Deep Reinforcement Learning (RL) systems is that they return an action no matter what situation they are confronted with. This is true even for situations that differ entirely from those an agent has been trained for. Although crucial in safety-critical applications, dealing with such situations is inherently difficult. Various approaches have been proposed in this direction, such as robustness, domain adaption, domain generalization, and out-of-distribution detection. In this work, we provide an overview of approaches towards the more general problem of dealing with disturbances to the environment of RL agents and show how they struggle to provide clear boundaries when mapped to safety-critical problems. To mitigate this, we propose to formalize the changes in the environment in terms of the Markov Decision Process (MDP), resulting in a more formal framework when dealing with such problems. We apply this framework to an example real-world scenario and show how it helps to isolate safety concerns.
  • Publication
    Assessing Box Merging Strategies and Uncertainty Estimation Methods in Multimodel Object Detection
    ( 2020)
    Schmoeller Roza, Felippe
    ;
    Henne, Maximilian
    ;
    Roscher, Karsten
    ;
    Günnemann, Stephan
    This paper examines the impact of different box merging strategies for sampling-based uncertainty estimation methods in object detection. Also, a comparison between the almost exclusively used softmax confidence scores and the predicted variances on the quality of the final predictions estimates is presented. The results suggest that estimated variances are a stronger predictor for the detection quality. However, variance-based merging strategies do not improve significantly over the confidence-based alternative for the given setup. In contrast, we show that different methods to estimate the uncertainty of the predictions have a significant influence on the quality of the ensembling outcome. Since mAP does not reward uncertainty estimates, such improvements were only noticeable on the resulting PDQ scores.
  • Publication
    CVIP: A Protocol for Complex Interactions Among Connected Vehicles
    ( 2020)
    Häfner, Bernhard
    ;
    Jiru, Josef
    ;
    Roscher, Karsten
    ;
    Ott, Jörg
    ;
    Schmitt, Georg A.
    ;
    Sevilmis, Yagmur
    Automated vehicles need to interact: to create mutual awareness and to coordinate maneuvers. How this interaction shall be achieved is still an open issue. Several new protocols are discussed for cooperative services such as changing lanes or overtaking, e.g., within the European Telecommunications Standards Institute (ETSI) and Society of Automotive Engineers (SAE). These communication protocols are, however, usually specific to individual maneuvers or based on implicit assumptions on other vehicles' intentions. To enable reuse and support extensibility towards future maneuvers, we propose CVIP, a protocol framework for complex vehicular interactions. CVIP supports explicitly negotiating maneuvers between the involved vehicles and allows monitoring maneuver progress via status updates. We present our design in detail and demonstrate via simulations that it enables complex inter-vehicle interactions in a flexible, efficient and robust manner. We also discuss open questions to be answered before complex interactions among automated vehicles can become a reality.