Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Deepcabac: Plug & Play Compression of Neural Network Weights and Weight Updates

: Neumann, D.; Sattler, F.; Kirchhoffer, H.; Wiedemann, S.; Müller, K.; Schwarz, H.; Wiegand, T.; Marpe, D.; Samek, W.


Institute of Electrical and Electronics Engineers -IEEE-; IEEE Signal Processing Society:
IEEE International Conference on Image Processing, ICIP 2020. Proceedings : September 25-28, 2020 Virtual Conference, Abu Dhabi, United Arab Emirates
Piscataway, NJ: IEEE, 2020
ISBN: 978-1-7281-6395-6
ISBN: 978-1-7281-6394-9
ISBN: 978-1-7281-6396-3
International Conference on Image Processing (ICIP) <2020, Online>
Fraunhofer HHI ()

An increasing number of distributed machine learning applications require efficient communication of neural network parameterizations. DeepCABAC, an algorithm in the current working draft of the emerging MPEG-7 part 17 standard for compression of neural networks for multimedia content description and analysis, has demonstrated high compression gains for a variety of neural network models. In this paper we propose a method for employing DeepCABAC in a Federated Learning scenario for the exchange of intermediate differential parameterizations. Furthermore, we discuss the efficiency of DeepCABAC when compressing trained neural networks. Our experiments on large neural networks show that in both scenarios, DeepCABAC achieves competitive compression rates, without degrading the network accuracy.