@article{Dom2021,Author = {Domenik Helms, Karl Amende, Saqib Bukhari, Thies de Graaff, Alexander Frickenstein, Frank Hafner, Tobias Hirscher, Sven Mantowsky, Georg Schneider, Manoj-Rohit Vemparala},Title = {Optimizing Neural Networks for Embedded Hardware},Journal = {Proceedings of the 2021 Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design},Year = {2021},Pages = {1-6},Url = {https://ieeexplore.ieee.org/document/9547911},type = {article},Abstract = {Neural networks are a pervasive technology, whichis, however, still held back in the area of embedded systems bythe high resource requirements, especially memory size, memoryaccess time and power dissipation. In recent years, severaldifferent methods have been proposed to transform given neuralnetworks in such a way that they can get by with much fewerresources while maintaining almost the same accuracy. This workreviews, categorizes and describes the state of the art in adaptingand simplifying neural networks to make them better applicableto embedded systems. Even though we developed this study froma purely automotive context, the techniques described are alsovalid in other areas.}}@COMMENT{Bibtex file generated on }