Reconfigurable FPGA implementation of neural networks
作者： Zbigniew Hajduk
DOI: 10.1016/J.NEUCOM.2018.04.077
关键词:
摘要: Abstract This brief paper presents two implementations of feed-forward artificial neural networks in FPGAs. The differ the FPGA resources requirement and calculations speed. Both exercise floating point arithmetic, apply very high accuracy activation function realization, enable easy alteration network's structure without need a re-implementation entire project.
sci-hub.se 参考文章(15)
Jacek Kluska, Zbigniew Hajduk, Hardware Implementation of P1-TS Fuzzy Rule-Based Systems on FPGA international conference on artificial intelligence and soft computing. pp. 282- 293 ,(2013) , 10.1007/978-3-642-38658-9_26
Vipin Tiwari, Nilay Khare, Hardware implementation of neural network with Sigmoidal activation functions using CORDIC Microprocessors and Microsystems. ,vol. 39, pp. 373- 381 ,(2015) , 10.1016/J.MICPRO.2015.05.012
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
Darío Baptista, Fernando Morgado-Dias, Low-resource hardware implementation of the hyperbolic tangent for artificial neural networks Neural Computing and Applications. ,vol. 23, pp. 601- 607 ,(2013) , 10.1007/S00521-013-1407-X
Zbigniew Hajduk, Bartosz Trybus, Jan Sadolewski, Architecture of FPGA Embedded Multiprocessor Programmable Controller IEEE Transactions on Industrial Electronics. ,vol. 62, pp. 2952- 2961 ,(2015) , 10.1109/TIE.2014.2362888
Vishnu P. Nambiar, Mohamed Khalil-Hani, Riadh Sahnoun, M.N. Marsono, Hardware implementation of evolvable block-based neural networks utilizing a cost efficient sigmoid-like activation function Neurocomputing. ,vol. 140, pp. 228- 241 ,(2014) , 10.1016/J.NEUCOM.2014.03.018
Janardan Misra, Indranil Saha, Artificial neural networks in hardware: A survey of two decades of progress Neurocomputing. ,vol. 74, pp. 239- 255 ,(2010) , 10.1016/J.NEUCOM.2010.03.021
Ivo Nascimento, Ricardo Jardim, Fernando Morgado-Dias, A new solution to the hyperbolic tangent implementation in hardware: polynomial modeling of the fractional exponential part Neural Computing and Applications. ,vol. 23, pp. 363- 369 ,(2013) , 10.1007/S00521-012-0919-0
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
Fernando Morgado Dias, Rui Borralho, Pedro Fontes, Ana Antunes, FTSET-a software tool for fault tolerance evaluation and improvement Neural Computing and Applications. ,vol. 19, pp. 701- 712 ,(2010) , 10.1007/S00521-009-0329-0