主 题:FPGA-based Acceleration for Deep Neural Networks
主 讲 人:关义金(北京大学)
时 间:2017年9月26日下午2:30
地 点:信息工程学院二层大会议室
主 办 单 位: 首都师范大学信息工程学院
Speaker Biology:
Yijin Guan is a PhD candidate in computer architecture research at Peking University, under the advisement of Prof. Guangyu Sun at PKU and Prof. Jason Cong at UCLA. His main research interests are deep learning architectures, FPGA-based accelerator designs and system designs.
Abstract:
Deep Neural Networks (DNNs) have demonstrated great success in numerous applications such as image classification, speech recognition, video analysis, etc. However, DNNs are much more computation-intensive and memory-intensive than previous shallow models. Thus, it is challenging to deploy DNNs in both large-scale data centers and real-time embedded systems. Considering performance, flexibility, and energy efficiency, FPGA-based accelerator for DNNs is a promising solution.
In this talk, I will introduce our representative works on FPGA-based acceleration for DNNs, including dedicated FPGA-based accelerators for specific DNN models, multi-board approaches for performance improvements, and automated design flows for mapping DNNs onto FPGAs.