Co-lead by Franz Pernkopf (Technical University of Graz) and Holger Fröning (ZITI)
Deep representation learning is one of the main factors for the recent performance boost in many image, signal and speech processing problems. This is particularly true when having big amounts of data and almost unlimited computing resources available. However, in real-world scenarios the computing infrastructure is often restricted and the computational requirements are not fulfilled.
In the DeepChip project, researchers from Technical University of Graz, Austria, and Ruprecht-Karls University of Heidelberg, Germany, have partnered to jointly explore algorithmic and architectural challenges when bringing the premise of Deep learning to resource-constrained processing like embedded systems and mobile chips. The key properties of interest are reducing the computational complexity of deep learning applications, and optimized performance in terms of execution time and energy consumption on embedded heterogeneous hardware while hiding heterogeneity from the user.
