S Tructured M Atrices in Fpga S
نویسندگان
چکیده
Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The recent pruning based work ESE (Han et al., 2017) suffers from degradation of performance/energy efficiency due to the irregular network structure after pruning. We propose block-circulant matrices for weight matrix representation in RNNs, thereby achieving simultaneous model compression and acceleration. We aim to implement RNNs in FPGA with highest performance and energy efficiency, with certain accuracy requirement (negligible accuracy degradation). Experimental results on actual FPGA deployments shows that the proposed framework achieves a maximum energy efficiency improvement of 35.7× compared with ESE.
منابع مشابه
Generation of Three-Phase PWM Inverter using Xilinx FPGA and its Application for Utility Connected PV System (RESEARCH NOTE)
Analysis and practical implementation of the regular symmetric sampled three-phase PWM inverter waveform has been presented in this paper. It is digitally implemented on a Xilinx field programmable gate array FPGA, and the essential considerations involved in the feasibility of using a Xilinx XC4008E software-based to generate PWM has been discussed. All the necessary Xilinx hardware/software t...
متن کاملFPGA-based of Thermogram Enhancement Algorithm for Non-destructive Thermal Characterization
متن کامل
طراحی و پیاده‌سازی رمزنگار AES در بستر FPGA برای خطوط پرسرعت
الگوریتم رمزنگاری AES یکی از متداول‌ترین الگوریتم‌های رمزنگاری متقارن است. به‌علت قابلیت‌های این الگوریتم، آن را می‌توان بر روی بسترهای مختلفی ازجمله بـر روی بسـترهای سخـت‌افزاری نظیر FPGA پیاده‌سازی کرد. همچنین به‌علت ساختار الگوریتم می‌توان مسیر داده را به‌صورت چرخشی و یا غیر چرخشی پیاده‌سازی نمود. ازآنجاکه بسته به کاربرد، استفاده از هریک از این دو معماری تأثیر فراوانی بر میزان گذردهی و میزان...
متن کاملFPGA Can be Implemented Using Advanced Encryption Standard Algorithm
This paper mainly focused on implementation of AES encryption and decryption standard AES-128. All the transformations of both Encryption and Decryption are simulated using an iterativedesign approach in order to minimize the hardware consumption. This method can make it avery low-complex architecture, especially in saving the hardware resource in implementing theAES InverseSub Bytes module and...
متن کامل