DUTh: Architectural Improvements for Low-Power and Functional Safety of Dataflow CNN Accelerators Using HLS
In this webinar, Dionysios Filippas from Democritus University of Thrace will focus on spatial dataflow alternatives, where CNNs are implemented using a series of dedicated convolutions engines, where the data are streamed from one layer to the other through custom memory buffers.
Deep Convolution Neural Networks (CNNs) are dominant in modern Machine Learning (ML) applications. Their acceleration directly in hardware calls for a multifaceted approach that combines high-performance, energy efficiency, and functional safety. These requirements hold for every CNN architecture, including both systolic and spatial dataflow alternatives. In this webinar, Dionysios Filippas from Democritus University of Thrace will focus on the latter case, where CNNs are implemented using a series of dedicated convolutions engines, where the data are streamed from one layer to the other through custom memory buffers.