AI Laboratory for System-level Design of ML-based Signal Processing Applications
The BMBF-founded project "KISS: AI Laboratory for System-level Design of ML-based Signal Processing Applications" is investigating new development tools to improve AI-based algorithms in signal processing and how to use computers to convert them into highly efficient implementations. The resulting reduction in development times promises improved or even completely new products and services for mobility, communication and entertainment.
All the research findings are being passed on to all relevant interested parties by means of university teaching and training in industry. This includes both the generation of training data as well as the efficient implementation and optimization of systems consisting of different computing units such as graphics cards or programmable chips. Combining traditional and machine learning methods increases the algorithms’ applicability and lowers the barriers to entry for industrial companies.
To achieve these goals, the algorithms to be applied are modeled with a high level of abstraction in form of semantic models, which can then be optimized using computers and transferred to different target platforms. The project team also provides concepts for how to obtain training data from existing data pools and simulation environments. To ensure that these methods are suited for practical use, they are tested on various video, speech and audio applications.
KISS is a joint project of Fraunhofer IIS and the Chair for Hardware/Software Co-Design at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). Further information about the project can be found here https://www.iis.fraunhofer.de/kiss
- Heidorn C., Witterauf M., Hannig F., Teich J.:
Efficient Mapping of CNNs onto Tightly Coupled Processor Arrays
In: Journal of Computers 14 (2019), p. 541-556
- Arvind TKR., Brand M., Heidorn C., Boppu S., Hannig F., Teich J.:
Hardware Implementation of Hyperbolic Tangent Activation Function for Floating Point Formats
24th International Symposium on VLSI Design and Test (VDAT) (Bhubaneswar, 23. July 2020 - 25. July 2020)
In: Proceedings of the 24th International Symposium on VLSI Design and Test (VDAT) 2020
- Heidorn C., Hannig F., Teich J.:
Design Space Exploration for Layer-parallel Execution of Convolutional Neural Networks on CGRAs
International Workshop on Software and Compilers for Embedded Systems (SCOPES) (St. Goar, 25. May 2020 - 26. May 2020)
In: Proceedings of the 23rd International Workshop on Software and Compilers for Embedded Systems (SCOPES) 2020