An Efficient Architecture for Floating Point Based MISO Neural Neworks on FPGA
作者： Antonino Laudani , Gabriele Maria Lozito , Francesco Riganti Fulginei , Alessandro Salvini
关键词:
摘要: The present paper documents the research towards development of an efficient algorithm to compute result from a multiple-input-single-output Neural Network using floating-point arithmetic on FPGA. proposed focus optimizing pipeline delays by splitting "Multiply and accumulate" into separate steps partial products. It is revisit classical for NN computation, able overcome main computation bottleneck in FPGA environment. can be implemented architecture that fully exploits performance blocks, thus allowing very fast neural network. presented as target Cyclone II Device.
ijssst.info参考文章(21)
Dario Baptista, Dias F. Morgado, On the Implementation of Different Hyperbolic Tangent Solutions in FPGA Controlo’2012. ,(2012)
Xi Chen, Gaofeng Wang, Wei Zhou, Sheng Chang, Shilei Sun, Efficient Sigmoid Function for Neural Networks Based FPGA Design Lecture Notes in Computer Science. pp. 672- 677 ,(2006) , 10.1007/11816157_80
Ayman Youssef, Karim Mohammed, Amin Nasar, A Reconfigurable, Generic and Programmable Feed Forward Neural Network Implementation in FPGA international conference on computer modelling and simulation. pp. 9- 13 ,(2012) , 10.1109/UKSIM.2012.12
F. R. FULGINEI, A. LAUDANI, A. SALVINI, M. PARODI, Automatic and Parallel Optimized Learning for Neural Networks performing MIMO Applications Advances in Electrical and Computer Engineering. ,vol. 13, pp. 3- 12 ,(2013) , 10.4316/AECE.2013.01001
Gabriele-Maria Lozito, Antonino Laudani, Francesco Riganti Fulginei, Alessandro Salvini, FPGA Implementations of Feed Forward Neural Network by using Floating Point Hardware Accelerators Advances in Electrical and Electronic Engineering. ,vol. 12, pp. 30- 39 ,(2014) , 10.15598/AEEE.V12I1.831
John D. Cappello, Dave Strenski, A practical measure of FPGA floating point acceleration for High Performance Computing application specific systems architectures and processors. pp. 160- 167 ,(2013) , 10.1109/ASAP.2013.6567570
Francesco Riganti Fulginei, Alessandro Salvini, Neural Network Approach for Modelling Hysteretic Magnetic Materials Under Distorted Excitations IEEE Transactions on Magnetics. ,vol. 48, pp. 307- 310 ,(2012) , 10.1109/TMAG.2011.2176106
Pedro Ferreira, Pedro Ribeiro, Ana Antunes, Fernando Morgado Dias, A high bit resolution FPGA implementation of a FNN with a new algorithm for the activation function Neurocomputing. ,vol. 71, pp. 71- 77 ,(2007) , 10.1016/J.NEUCOM.2006.11.028
A. Salvini, C. Coltelli, Prediction of dynamic hysteresis under highly distorted exciting fields by neural networks and actual frequency transplantation IEEE Transactions on Magnetics. ,vol. 37, pp. 3315- 3319 ,(2001) , 10.1109/20.952603
Ramazan Bayindir, Alper Gorgun, Hardware Implementation of a Real-Time Neural Network Controller Set for Reactive Power Compensation Systems international conference on machine learning and applications. pp. 699- 703 ,(2010) , 10.1109/ICMLA.2010.107