Skip to main content

3 Result(s)

within Tobias Schuster Remove this filter Computer Systems Organization and Communication Networks Remove this filter

and

  1. No Access

    Chapter and Conference Paper

    Logically Sound Arguments for the Effectiveness of ML Safety Measures

    We investigate the issues of achieving sufficient rigor in the arguments for the safety of machine learning functions. By considering the known weaknesses of DNN-based 2D bounding box detection algorithms, we ...

    Chih-Hong Cheng, Tobias Schuster… in Computer Safety, Reliability, and Securit… (2022)

  2. No Access

    Chapter and Conference Paper

    Formally Compensating Performance Limitations for Imprecise 2D Object Detection

    In this paper, we consider the imperfection within machine learning-based 2D object detection and its impact on safety. We address a special sub-type of performance limitations related to the misalignment of b...

    Tobias Schuster, Emmanouil Seferis… in Computer Safety, Reliability, and Security (2022)

  3. No Access

    Chapter and Conference Paper

    Neural Criticality Metric for Object Detection Deep Neural Networks

    The complexity of state-of-the-art Deep Neural Network (DNN) architectures exacerbates the search for safety relevant metrics and methods that could be used for functional safety assessments. In this article, ...

    Václav Diviš, Tobias Schuster, Marek Hrúz in Computer Safety, Reliability, and Securit… (2022)