The“memory wall”of traditional von Neumann computing systems severely restricts the efficiency of data-intensive task execution,while in-memory computing(IMC)architecture is a promising approach to breaking the bott...The“memory wall”of traditional von Neumann computing systems severely restricts the efficiency of data-intensive task execution,while in-memory computing(IMC)architecture is a promising approach to breaking the bottleneck.Although variations and instability in ultra-scaled memory cells seriously degrade the calculation accuracy in IMC architectures,stochastic computing(SC)can compensate for these shortcomings due to its low sensitivity to cell disturbances.Furthermore,massive parallel computing can be processed to improve the speed and efficiency of the system.In this paper,by designing logic functions in NOR flash arrays,SC in IMC for the image edge detection is realized,demonstrating ultra-low computational complexity and power consumption(25.5 fJ/pixel at 2-bit sequence length).More impressively,the noise immunity is 6 times higher than that of the traditional binary method,showing good tolerances to cell variation and reliability degradation when implementing massive parallel computation in the array.展开更多
Rapid‐Single‐Flux‐Quantum(RSFQ)circuit technology is well known for its low power consumption and latency,which enables digital signal processing up to tens of GHz.As a fundamental digital filter,the Finite Impulse...Rapid‐Single‐Flux‐Quantum(RSFQ)circuit technology is well known for its low power consumption and latency,which enables digital signal processing up to tens of GHz.As a fundamental digital filter,the Finite Impulse Response(FIR)filter has wide applications in communication systems.A design of an FIR filter based on RSFQ circuit technology is proposed.However,the FIR filter consumes large amounts of adders and multipliers.Based on Stochastic Computing(SC)theory with which adder and multiplier are much simpler,the hardware cost of FIR filter is dramatically reduced.A novel stochastic number generator(SNG),a stochastic‐tobinary converter(SBC),and the FIR filter were designed and verified via logic simulation with a target frequency of 10 GHz.The results indicated the FIR filter performs correct operations.The proposed FIR filter consists of 2255 Josephson junctions(JJs)without wiring cells(i.e.,Josephson Transmission Lines(JTLs),Passive Transmission Lines(PTLs)),which is acceptable,making it possible to be used in RSFQ digital signal processors.展开更多
Topic modeling is a mainstream and effective technology to deal with text data, with wide applications in text analysis, natural language, personalized recommendation, computer vision, etc. Among all the known topic m...Topic modeling is a mainstream and effective technology to deal with text data, with wide applications in text analysis, natural language, personalized recommendation, computer vision, etc. Among all the known topic models, supervised Latent Dirichlet Allocation (sLDA) is acknowledged as a popular and competitive supervised topic model. How- ever, the gradual increase of the scale of datasets makes sLDA more and more inefficient and time-consuming, and limits its applications in a very narrow range. To solve it, a parallel online sLDA, named PO-sLDA (Parallel and Online sLDA), is proposed in this study. It uses the stochastic variational inference as the learning method to make the training procedure more rapid and efficient, and a parallel computing mechanism implemented via the MapReduce framework is proposed to promote the capacity of cloud computing and big data processing. The online training capacity supported by PO-sLDA expands the application scope of this approach, making it instrumental for real-life applications with high real-time demand. The validation using two datasets with different sizes shows that the proposed approach has the comparative accuracy as the sLDA and can efficiently accelerate the training procedure. Moreover, its good convergence and online training capacity make it lucrative for the large-scale text data analyzing and processing.展开更多
To improve the performance of extended particle swarm optimizer, a novel means of stochastic weight deployment is proposed for the iterative equation of velocity updation. In this scheme, one of the weights is specifi...To improve the performance of extended particle swarm optimizer, a novel means of stochastic weight deployment is proposed for the iterative equation of velocity updation. In this scheme, one of the weights is specified to a random number within the range of [0, 1] and the other two remain constant configurations. The simulations show that this weight strategy outperforms the previous deterministic approach with respect to success rate and convergence speed. The experiments also reveal that if the weight for global best neighbor is specified to a stochastic number, extended particle swarm optimizer achieves high and robust performance on the given multi-modal function.展开更多
基金supported by the National Natural Science Foundation of China(Nos.62034006,91964105,61874068)the China Key Research and Development Program(No.2016YFA0201802)+1 种基金the Natural Science Foundation of Shandong Province(No.ZR2020JQ28)Program of Qilu Young Scholars of Shandong University。
文摘The“memory wall”of traditional von Neumann computing systems severely restricts the efficiency of data-intensive task execution,while in-memory computing(IMC)architecture is a promising approach to breaking the bottleneck.Although variations and instability in ultra-scaled memory cells seriously degrade the calculation accuracy in IMC architectures,stochastic computing(SC)can compensate for these shortcomings due to its low sensitivity to cell disturbances.Furthermore,massive parallel computing can be processed to improve the speed and efficiency of the system.In this paper,by designing logic functions in NOR flash arrays,SC in IMC for the image edge detection is realized,demonstrating ultra-low computational complexity and power consumption(25.5 fJ/pixel at 2-bit sequence length).More impressively,the noise immunity is 6 times higher than that of the traditional binary method,showing good tolerances to cell variation and reliability degradation when implementing massive parallel computation in the array.
基金supported in part by the Strategic Priority Research Program of Chinese Academy of Sciences,under Grant XDA18000000.
文摘Rapid‐Single‐Flux‐Quantum(RSFQ)circuit technology is well known for its low power consumption and latency,which enables digital signal processing up to tens of GHz.As a fundamental digital filter,the Finite Impulse Response(FIR)filter has wide applications in communication systems.A design of an FIR filter based on RSFQ circuit technology is proposed.However,the FIR filter consumes large amounts of adders and multipliers.Based on Stochastic Computing(SC)theory with which adder and multiplier are much simpler,the hardware cost of FIR filter is dramatically reduced.A novel stochastic number generator(SNG),a stochastic‐tobinary converter(SBC),and the FIR filter were designed and verified via logic simulation with a target frequency of 10 GHz.The results indicated the FIR filter performs correct operations.The proposed FIR filter consists of 2255 Josephson junctions(JJs)without wiring cells(i.e.,Josephson Transmission Lines(JTLs),Passive Transmission Lines(PTLs)),which is acceptable,making it possible to be used in RSFQ digital signal processors.
基金This work was supported in part by the National Natural Science Foundation of China under Grant Nos. 61572226 and 61876069, and the Key Scientific and Technological Research and Development Project of Jilin Province of China under Grant Nos. 20180201067GX and 20180201044GX.
文摘Topic modeling is a mainstream and effective technology to deal with text data, with wide applications in text analysis, natural language, personalized recommendation, computer vision, etc. Among all the known topic models, supervised Latent Dirichlet Allocation (sLDA) is acknowledged as a popular and competitive supervised topic model. How- ever, the gradual increase of the scale of datasets makes sLDA more and more inefficient and time-consuming, and limits its applications in a very narrow range. To solve it, a parallel online sLDA, named PO-sLDA (Parallel and Online sLDA), is proposed in this study. It uses the stochastic variational inference as the learning method to make the training procedure more rapid and efficient, and a parallel computing mechanism implemented via the MapReduce framework is proposed to promote the capacity of cloud computing and big data processing. The online training capacity supported by PO-sLDA expands the application scope of this approach, making it instrumental for real-life applications with high real-time demand. The validation using two datasets with different sizes shows that the proposed approach has the comparative accuracy as the sLDA and can efficiently accelerate the training procedure. Moreover, its good convergence and online training capacity make it lucrative for the large-scale text data analyzing and processing.
基金the Natural Science Foundation of the Anhui Higher Education Institutions (KJ2008B151)Key Laboratory of Information Management and Information Economics, Ministry of Education (F0607-36)
文摘To improve the performance of extended particle swarm optimizer, a novel means of stochastic weight deployment is proposed for the iterative equation of velocity updation. In this scheme, one of the weights is specified to a random number within the range of [0, 1] and the other two remain constant configurations. The simulations show that this weight strategy outperforms the previous deterministic approach with respect to success rate and convergence speed. The experiments also reveal that if the weight for global best neighbor is specified to a stochastic number, extended particle swarm optimizer achieves high and robust performance on the given multi-modal function.