no code implementations • 30 Jan 2024 • Jens Henriksson, Christian Berger, Stig Ursing, Markus Borg
Safety measures need to be systemically investigated to what extent they evaluate the intended performance of Deep Neural Networks (DNNs) for critical applications.
no code implementations • 26 Apr 2022 • Jens Henriksson, Christian Berger, Markus Borg, Lars Tornberg, Sankar Raman Sathyamoorthy, Cristofer Englund
Implementing Deep Neural Networks (DNN) for non-safety related applications have shown remarkable achievements over the past years; however, for using DNNs in safety critical applications, we are missing approaches for verifying the robustness of such models.
no code implementations • 26 Apr 2022 • Jens Henriksson, Christian Berger, Stig Ursing
As the NN is trained on well annotated images, in this paper we study the variations of confidence levels from the NN when tested on hand-crafted occlusion added to a test set.
1 code implementation • 16 Apr 2022 • Markus Borg, Jens Henriksson, Kasper Socha, Olof Lennartsson, Elias Sonnsjö Lönegren, Thanh Bui, Piotr Tomaszewski, Sankar Raman Sathyamoorthy, Sebastian Brink, Mahshid Helali Moghadam
We initiated a research project with the goal to demonstrate a complete safety case for an ML component in an open automotive system.
no code implementations • 29 Mar 2021 • Jens Henriksson, Christian Berger, Markus Borg, Lars Tornberg, Sankar Raman Sathyamoorthy, Cristofer Englund
Understanding the relationship between training results and supervisor performance is valuable to improve robustness of the model and indicates where more work has to be done to create generalized models for safety critical applications.
no code implementations • 16 Feb 2020 • Dhasarathy Parthasarathy, Karl Bäckström, Jens Henriksson, Sólrún Einarsdóttir
Testing automotive mechatronic systems partly uses the software-in-the-loop approach, where systematically covering inputs of the system-under-test remains a major challenge.
no code implementations • 4 Mar 2019 • Jens Henriksson, Christian Berger, Markus Borg, Lars Tornberg, Cristofer Englund, Sankar Raman Sathyamoorthy, Stig Ursing
Deep Neural Networks (DNN) have improved the quality of several non-safety related products in the past years.