Field-programmable gate arrays (FPGAs) were invented by Xilinx in 1985, ie, about 35 years
ago. Currently, they are part of heterogeneous computer platforms that combine different …

… In this paper we present pvFPGA, the first system design solution for virtualizing an FPGA-based
hardware accelerator on the x86 platform. Our design adopts the Xen virtual machine …

… Mittal, “A survey of FPGA-based accelerators for convolutional neural networks,” Neural …
Fanucci, “A FPGA-based hardware accelerator for CNNs using on-chip memories only: …

… the FPGA-based systems. To further emphasize the need for designing FPGA-based
approximate … the efficiency of both ASIC-based and FPGA-based implementations of state-of-the-art …

… over other types of hardware accelerators. … FPGA based deep neural network accelerators
from a few important aspects of hardware design such as algorithmic optimization, hardware …

… directly synthesized for the FPGA-based systems. To further emphasize the need for
designing FPGA-based approximate modules, we present the following motivational case study. …

… So ware design automation tries to run di erent networks on the same hardware accelerator
by simply changing the input, in most cases, an instruction sequence. e di erence between …

… acceleration. The techniques investigated in this paper represent the recent trends in FPGA-based
accelerators of … hardware accelerators and to be useful for deep learning researchers. …

… movements and to heterogeneous computing platforms that integrate dedicated and
specialized hardware accelerators. The performance of battery-driven mobile devices we …

… multicore-processing with hardware accelerators, have enabled … , we present an FPGA-based
hardware accelerator for the … in detail our FPGA-based hardware accelerator, whose block …