An Energy-Efficient Architecture for Binary Weight Convolutional Neural Networks
Abstract: Binary weight convolutional neural networks (BCNNs) can achieve near state-of-the-art classification accuracy and have far less computation complexity compared with traditional CNNs using high-precision weights. Due to their binary weights, BCNNs are well suited for vision-based Internet-of-Things systems being sensitive to power consumption. BCNNs make it possible to achieve very high throughput with moderate power dissipation. In this paper, an energy-efficient architecture for BCNNs is proposed. It fully exploits the binary weights and other hardware-friendly characteristics of BCNNs. A judicious processing schedule is proposed so that off-chip I/O access is minimized and activations are maximally reused. To significantly reduce the critical path delay, we introduce optimized compressor trees and approximate binary multipliers with two novel compensation schemes. The latter is able to save significant hardware resource, and almost no computation accuracy is compromised. Taking advantage of error resiliency of BCNNs, an innovative approximate adder is developed, which significantly reduces the silicon area and data path delay. Thorough error analysis and extensive experimental results on several data sets show that the approximate adders in the data path cause negligible accuracy loss. Moreover, algorithmic transformations for certain layers of BCNNs and a memory-efficient quantization scheme are incorporated to further reduce the energy cost and on-chip storage requirement. Finally, the proposed BCNN hardware architecture is implemented with the SMIC 130-nm technology. The postlayout results demonstrate that our design can achieve an energy efficiency over 2.0TOp/s/W when scaled to 65 nm, which is more than two times better than the prior art.
VLSI Projects,IEEE VLSI Projects,latest vlsi projects,2018 VLSI Projects,VLSI Projects in Bangalore,VLSI projects institutes in bangalore,VLSI live projects in bangalore,VLSI academic projects,VLSI project centres,M.Tech VLSI projects in bangalore,M Tech VLSI projects institutes in bangalore,FPGA projects in bangalore,ieee vlsi,vlsi ieee papers,mtech vlsi,fpga projects using vhdl,mini project on image processing,vlsi paper,vlsi ieee papers,ieee project papers,vlsi institutes in bangalore,ofdm projects,vlsi projects using vhdl,projects based on digital signal processing,vhdl based projects,latest vlsi projects,vlsi project institutes in bangalore,VLSI Project,vlsiproject,vlsi project institute in bangalore,vlsi project idea,idea in vlsiproject,idea in vlsiprojects,idea in vlsi project,idea in vlsiprojects,M.Tech VLSI Projects in Bangalore,M.Tech FPGA Projects in Bangalore,ECE VLSI Projects in Bangalore,VLSI Academic Projects in Bangalore,VLSI Live Projects in Bangalore,VLSI Real Time Projects in Bangalore,VLSI Projects for MTech 2018,VLSI Projects for MTech 2018,VLSI Projects for MTech in Bangalore,FPGA based Projects for M.Tech,download 2018 VLSI Project list,VLSI project centre,VLSI academic projects,vlsi ieee papers,new vlsi projects,mtech vlsi,fpga projects using vhdl,mini project on image processing,vlsi paper,vlsi ieee papers,ieee project papers,vlsi institutes in bangalore,ofdm projects,vlsi projects using vhdl,projects based on digital signal processing,vhdl based projects,latest vlsi projects,vlsi project institute in bangalore,vlsi project idea,idea in vlsiproject,idea in vlsiprojects,idea in vlsi project,idea in vlsiprojects
del.icio.us
Google