Work

An enegy-efficient FPGA accelerator for convolutional neural networks

Year: 2018
Type: dissertation
Abstract: This project focuses on a state-of-the-art DNN specifically build for image clas sification. We develop a new architecture design that will run on an experimental Intel accelerator platform called HARP. We obtain low-power solution and evaluate the possibilities and requierements of HARPplatform.
Language: English
Open Access status: green
OpenAlex ID: W2921314358
Cites: 6
Cited by:
Related to: 10