TY  - CHAP
U1  - Teil eines Buches
A1  - Houben, Sebastian
A1  - Abrecht, Stephanie
A1  - Akila, Maram
A1  - Bär, Andreas
A1  - Brockherde, Felix
A1  - Feifel, Patrick
A1  - Fingscheidt, Tim
A1  - Gannamaneni, Sujan Sai
A1  - Ghobadi, Seyed Eghbal
A1  - Hammam, Ahmed
A1  - Haselhoff, Anselm
A1  - Hauser, Felix
A1  - Heinzemann, Christian
A1  - Hoffmann, Marco
A1  - Kapoor, Nikhil
A1  - Kappel, Falk
A1  - Klingner, Marvin
A1  - Kronenberger, Jan
A1  - Küppers, Fabian
A1  - Löhdefink, Jonas
A1  - Mlynarski, Michael
A1  - Mock, Michael
A1  - Mualla, Firas
A1  - Pavlitskaya, Svetlana
A1  - Poretschkin, Maximilian
A1  - Pohl, Alexander
A1  - Ravi-Kumar, Varun
A1  - Rosenzweig, Julia
A1  - Rottmann, Matthias
A1  - Rüping, Stefan
A1  - Sämann, Timo
A1  - Schneider, Jan David
A1  - Schulz, Elena
A1  - Schwalbe, Gesina
A1  - Sicking, Joachim
A1  - Srivastava, Toshika
A1  - Varghese, Serin
A1  - Weber, Michael
A1  - Wirkert, Sebastian
A1  - Wirtz, Tim
A1  - Woehrle, Matthias
T1  - Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
T2  - Fingscheidt, Gottschalk et al. (Hg.): Deep Neural Networks and Data for Automated Driving. Robustness, Uncertainty Quantification, and Insights Towards Safety
N2  - Deployment of modern data-driven machine learning methods, most often realized by deep neural networks (DNNs), in safety-critical applications such as health care, industrial plant control, or autonomous driving is highly challenging due to numerous model-inherent shortcomings. These shortcomings are diverse and range from a lack of generalization over insufficient interpretability and implausible predictions to directed attacks by means of malicious inputs. Cyber-physical systems employing DNNs are therefore likely to suffer from so-called safety concerns, properties that preclude their deployment as no argument or experimental setup can help to assess the remaining risk. In recent years, an abundance of state-of-the-art techniques aiming to address these safety concerns has emerged. This chapter provides a structured and broad overview of them. We first identify categories of insufficiencies to then describe research activities aiming at their detection, quantification, or mitigation. Our work addresses machine learning experts and safety engineers alike: The former ones might profit from the broad range of machine learning topics covered and discussions on limitations of recent methods. The latter ones might gain insights into the specifics of modern machine learning methods. We hope that this contribution fuels discussions on desiderata for machine learning systems and strategies on how to help to advance existing approaches accordingly.
Y1  - 2022
UN  - https://nbn-resolving.org/urn:nbn:de:hbz:1044-opus-65483
SN  - 978-3-031-01232-7
SB  - 978-3-031-01232-7
U6  - https://doi.org/10.1007/978-3-031-01233-4_1
DO  - https://doi.org/10.1007/978-3-031-01233-4_1
SP  - 3
EP  - 78
S1  - 76
PB  - Springer International Publishing AG
CY  - Cham
ER  -