Paper accepted at DATE 2020

We have received acceptance for a paper entitled "OrthrusPE: Runtime Reconfigurable Processing Elements for Binary Neural Networks", authored by Nael Fasfous, Manoj-Rohit Vemparala, Alexander Frickenstein, and Walter Stechele. In this paper, the authors propose a runtime reconfigurable processing element for mixed binary and fixed point arithmetic neural networks, based on Xilinx DSPs, optimizing throughput, resource utilization, and power efficiency.

The Design Automation and Test in Europe Conference (DATE) is the main European event for electronic design automation. DATE 2020 will take place in Grenoble, France, from March 09 to 13, 2020.