TY - CONF AB - Artificial neural networks in general and deep learning networks in particular established themselves as popular and powerful machine learning algorithms. While the often tremendous sizes of these networks are beneficial when solving complex tasks, the tremendous number of parameters also causes such networks to be vulnerable to malicious behavior such as adversarial perturbations. These perturbations can change a model's classification decision. Moreover, while single-step adversaries can easily be transferred from network to network, the transfer of more powerful multi-step adversaries has - usually - been rather difficult.In this work, we introduce a method for generating strong adversaries that can easily (and frequently) be transferred between different models. This method is then used to generate a large set of adversaries, based on which the effects of selected defense methods are experimentally assessed. At last, we introduce a novel, simple, yet effective approach to enhance the resilience of neural networks against adversaries and benchmark it against established defense methods. In contrast to the already existing methods, our proposed defense approach is much more efficient as it only requires a single additional forward-pass to achieve comparable performance results. AU - Seiler, Moritz Vinzent AU - Trautmann, Heike AU - Kerschke, Pascal ID - 46331 T2 - Proceedings of the International Joint Conference on Neural Networks (IJCNN) TI - Enhancing Resilience of Deep Learning Networks By Means of Transferable Adversaries ER - TY - GEN AU - Terfort, Tobias ID - 18639 TI - Enhancing Security by Usage of Universal One-Way Hash Functions ER - TY - JOUR AU - Grimminger-Seidensticker, Elke AU - Möhwald, Aiko Julia ID - 28341 IS - 3 JF - Physical Education and Sport Pedagogy SN - 1740-8989 TI - Enhancing social cohesion in PE classes within an intercultural learning program: results of a quasi-experimental intervention study VL - 25 ER - TY - CHAP AU - Haak, Inka AU - Gildehaus, Lara AU - Liebendörfer, Michael ED - Siller, Hans-Stefan ED - Weigel, Wolfgang ED - Wörler, Jan Franz ID - 37491 T2 - Beiträge zum Mathematikunterricht 2020 TI - Entstehung und Bedeutung von Lerngruppen in der Studieneingangsphase ER - TY - THES AU - Pietsch, Tommy ID - 42755 SN - 978-3-8440-7128-3 TI - Entwicklung des Prägeelementschweißens für Aluminium-Stahl-Verbindungen im Karosseriebau ER - TY - CONF AU - Kullmer, Gunter AU - Weiß, Deborah AU - Bauer, Benjamin AU - Richard, Hans Albert ID - 24012 TI - Entwicklung einer Axialrissprobe zur Ermittlung von bruchmechanischen Kennwerten für Rohre VL - DVM-Bericht 252 ER - TY - THES AU - Gerkens, Michael ID - 42758 SN - 978-3-8440-7583-0 TI - Entwicklung einer Methodik zur numerischen Simulation des Hochgeschwindigkeits-Bolzensetzens ER - TY - JOUR AU - Moritzer, Elmar AU - Krassmann, Dimitri ID - 24234 JF - Joining Plastics TI - Entwicklung einer neuartigen Fügetechnik für Organoblech-Hybridverbindungen ER - TY - THES AU - Resonnek, Verena ID - 27646 TI - Entwicklung einer Zylindertemperatureinstellungsregelung auf Basis von Fuzzy-Logik ER - TY - JOUR AU - Bauer, Anna AU - Reinhold, Peter AU - Sacher, Marc ID - 24956 JF - Phydid B, Didaktik der Physik, Beiträge zur DPG-Frühjahrstagung TI - Entwicklung eines Bewertungsmodells zur handlungsorientierten Messung experimenteller Kompetenz (Physik)Studierender ER -