With the continuous development of deep learning,Deep Convolutional Neural Network(DCNN)has attracted wide attention in the industry due to its high accuracy in image classification.Compared with other DCNN hard-ware ...With the continuous development of deep learning,Deep Convolutional Neural Network(DCNN)has attracted wide attention in the industry due to its high accuracy in image classification.Compared with other DCNN hard-ware deployment platforms,Field Programmable Gate Array(FPGA)has the advantages of being programmable,low power consumption,parallelism,and low cost.However,the enormous amount of calculation of DCNN and the limited logic capacity of FPGA restrict the energy efficiency of the DCNN accelerator.The traditional sequential sliding window method can improve the throughput of the DCNN accelerator by data multiplexing,but this method’s data multiplexing rate is low because it repeatedly reads the data between rows.This paper proposes a fast data readout strategy via the circular sliding window data reading method,it can improve the multiplexing rate of data between rows by optimizing the memory access order of input data.In addition,the multiplication bit width of the DCNN accelerator is much smaller than that of the Digital Signal Processing(DSP)on the FPGA,which means that there will be a waste of resources if a multiplication uses a single DSP.A multiplier sharing strategy is proposed,the multiplier of the accelerator is customized so that a single DSP block can complete multiple groups of 4,6,and 8-bit signed multiplication in parallel.Finally,based on two strategies of appeal,an FPGA optimized accelerator is proposed.The accelerator is customized by Verilog language and deployed on Xilinx VCU118.When the accelerator recognizes the CIRFAR-10 dataset,its energy efficiency is 39.98 GOPS/W,which provides 1.73×speedup energy efficiency over previous DCNN FPGA accelerators.When the accelerator recognizes the IMAGENET dataset,its energy efficiency is 41.12 GOPS/W,which shows 1.28×−3.14×energy efficiency compared with others.展开更多
The traditional manner to design public transportation system is to sequentially design the transit network and public bicycle network. A new public transportation system design problem that simultaneously considers b...The traditional manner to design public transportation system is to sequentially design the transit network and public bicycle network. A new public transportation system design problem that simultaneously considers both bus network design and public bicycle network design is proposed. The chemical reaction optimization(CRO) is designed to solve the problem. A shortcoming of CRO is that, when the two-molecule collisions take place, the molecules are randomly picked from the container.Hence, we improve CRO by employing different mating strategies. The computational results confirm the benefits of the mating strategies. Numerical experiments are conducted on the Sioux-Falls network. A comparison with the traditional sequential modeling framework indicates that the proposed approach has a better performance and is more robust. The practical applicability of the approach is proved by employing a real size network.展开更多
The quickly growing development and fierce competition in technical industry make Apps for The Future Inc. (AFF) has risk of losing market share and their customers' trust. Therefore, after the examine of three so...The quickly growing development and fierce competition in technical industry make Apps for The Future Inc. (AFF) has risk of losing market share and their customers' trust. Therefore, after the examine of three solutions which used by Nokia, Motorola and China Unicom based on the selected criteria, the suitable solutions that the AFF managers should follow are creating new flagship product with low price, finding development direction and keep investing. Those solutions could allow companies to keep creating, keep motivating and care their customers, and allow this company to be successful and regain customer trust.展开更多
Under the huge weight of the global financial crisis,the tire industry plunged into a depression for a quite long time.But Yokohama Rubber (China) Co.,Ltd.(Yokohama Rubber) has successfully run its business to survive...Under the huge weight of the global financial crisis,the tire industry plunged into a depression for a quite long time.But Yokohama Rubber (China) Co.,Ltd.(Yokohama Rubber) has successfully run its business to survive the crisis.展开更多
基金supported in part by the Major Program of the Ministry of Science and Technology of China under Grant 2019YFB2205102in part by the National Natural Science Foundation of China under Grant 61974164,62074166,61804181,62004219,62004220,62104256.
文摘With the continuous development of deep learning,Deep Convolutional Neural Network(DCNN)has attracted wide attention in the industry due to its high accuracy in image classification.Compared with other DCNN hard-ware deployment platforms,Field Programmable Gate Array(FPGA)has the advantages of being programmable,low power consumption,parallelism,and low cost.However,the enormous amount of calculation of DCNN and the limited logic capacity of FPGA restrict the energy efficiency of the DCNN accelerator.The traditional sequential sliding window method can improve the throughput of the DCNN accelerator by data multiplexing,but this method’s data multiplexing rate is low because it repeatedly reads the data between rows.This paper proposes a fast data readout strategy via the circular sliding window data reading method,it can improve the multiplexing rate of data between rows by optimizing the memory access order of input data.In addition,the multiplication bit width of the DCNN accelerator is much smaller than that of the Digital Signal Processing(DSP)on the FPGA,which means that there will be a waste of resources if a multiplication uses a single DSP.A multiplier sharing strategy is proposed,the multiplier of the accelerator is customized so that a single DSP block can complete multiple groups of 4,6,and 8-bit signed multiplication in parallel.Finally,based on two strategies of appeal,an FPGA optimized accelerator is proposed.The accelerator is customized by Verilog language and deployed on Xilinx VCU118.When the accelerator recognizes the CIRFAR-10 dataset,its energy efficiency is 39.98 GOPS/W,which provides 1.73×speedup energy efficiency over previous DCNN FPGA accelerators.When the accelerator recognizes the IMAGENET dataset,its energy efficiency is 41.12 GOPS/W,which shows 1.28×−3.14×energy efficiency compared with others.
基金Projects(71301115,71271150,71101102)supported by the National Natural Science Foundation of ChinaProject(20130032120009)supported by Specialized Research Fund for the Doctoral Program of Higher Education of China
文摘The traditional manner to design public transportation system is to sequentially design the transit network and public bicycle network. A new public transportation system design problem that simultaneously considers both bus network design and public bicycle network design is proposed. The chemical reaction optimization(CRO) is designed to solve the problem. A shortcoming of CRO is that, when the two-molecule collisions take place, the molecules are randomly picked from the container.Hence, we improve CRO by employing different mating strategies. The computational results confirm the benefits of the mating strategies. Numerical experiments are conducted on the Sioux-Falls network. A comparison with the traditional sequential modeling framework indicates that the proposed approach has a better performance and is more robust. The practical applicability of the approach is proved by employing a real size network.
文摘The quickly growing development and fierce competition in technical industry make Apps for The Future Inc. (AFF) has risk of losing market share and their customers' trust. Therefore, after the examine of three solutions which used by Nokia, Motorola and China Unicom based on the selected criteria, the suitable solutions that the AFF managers should follow are creating new flagship product with low price, finding development direction and keep investing. Those solutions could allow companies to keep creating, keep motivating and care their customers, and allow this company to be successful and regain customer trust.
文摘Under the huge weight of the global financial crisis,the tire industry plunged into a depression for a quite long time.But Yokohama Rubber (China) Co.,Ltd.(Yokohama Rubber) has successfully run its business to survive the crisis.