Efficient Ensemble Machine Learning Implementation on FPGA Using Partial Reconfiguration

  • Gian Carlo Cardarilli
  • Luca Di NunzioEmail author
  • Rocco Fazzolari
  • Daniele Giardino
  • Marco Matta
  • Marco Re
  • Francesca Silvestri
  • Sergio Spanò
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 573)


Ensemble Machine Learning (EML) consists of the combination of multiple Artificial Intelligence algorithms. This paper presents an efficient FPGA implementation of an Ensemble based on Long Short-Term Memory Networks (LSTM). For an efficient implementation, the proposed design uses the Partial Reconfiguration function available for FPGAs. Results are presented in terms of resources utilization, reconfiguration speed, power consumption and maximum clock frequency.



The authors would like to thank Xilinx Inc, for providing FPGA hardware and software tools by Xilinx University Program.


  1. 1.
    Lo Sciuto, G., Susi, G., Cammarata e, G., Capizzi, G.: A spiking neural network-based model for anaerobic digestion process. In: IEEE 23rd International Symposium on Power Electronics, Electrical Drives, Automation and Motion (SPEEDAM) (2016)Google Scholar
  2. 2.
    Brusca, S., Capizzi, G., Lo Sciuto e, G., Susi, G.: A new design methodology to predict wind farm energy production by means of a spiking neural network based-system. Int. J. Numer. Model. Electron. Netw. Devices Fields 7 (2017)Google Scholar
  3. 3.
    Scarpato, N., Pieroni, A., Di Nunzio, L., Fallucchi, F.: E-health-IoT universe: a review. Int. J. Adv. Sci. Eng. Inf. Technol. 7(6), 2328–2336 (2017)CrossRefGoogle Scholar
  4. 4.
    Cardarilli, G.C., Cristini, A., Di Nunzio, L., Re, M., Salerno, M., Susi, G.: Spiking neural networks based on LIF with latency: simulation and synchronization effects. In: Asilomar Conference on Signals, Systems and Computers, pp. 1838–1842 (2013)Google Scholar
  5. 5.
    Khanal, G.M., Acciarito, S., Cardarilli, G.C., Chakraborty, A., Di Nunzio, L., Fazzolari, R., Cristini, A., Re, M., Susi, G.: Synaptic behaviour in ZnO-rGO composites thin film memristor. Electron. Lett. 53(5), 296–298 (2017)CrossRefGoogle Scholar
  6. 6.
    Acciarito, S., Cardarilli, G.C., Cristini, A., Nunzio, L.D., Fazzolari, R., Khanal, G.M., Re, M., Susi, G.: Hardware design of LIF with Latency neuron model with memristive STDP synapses. Integr. VLSI J. 59, 81–89 (2017)CrossRefGoogle Scholar
  7. 7.
    Khanal, G.M., Cardarilli, G., Chakraborty, A., Acciarito, S., Mulla, M.Y., Di Nunzio, L., Fazzolari, R., Re, M.: A ZnO-rGO composite thin film discrete memristor. IEEE, ICSE, art. no. 7573608, pp. 129–132 (2016)Google Scholar
  8. 8.
    Acciarito, S., Cristini, A., Di Nunzio, L., Khanal, G.M., Susi, G.: An a VLSI driving circuit for memristor-based STDP. PRIME 2016, art. no. 7519503 (2016)Google Scholar
  9. 9.
    Opitz, D.; Maclin, R.: Popular ensemble methods: an empirical study. J. Artif. Intell. Res. 11, 169–198CrossRefGoogle Scholar
  10. 10.
    Polikar, R: Ensemble based systems in decision making. IEEE Circuits Syst. Mag. 6(3), 21–45CrossRefGoogle Scholar
  11. 11.
    Rokach, L: Ensemble-based classifiers. Artif. Intell. Rev. 33(1–2), 1–39MathSciNetCrossRefGoogle Scholar
  12. 12.
    Dalmasso, I., Galletti, I., Giuliano, R., Mazzenga, F.: WiMAX Networks for Emergency Management Based on UAVs. In: IEEE–AESS European Conference on Satellite Telecommunications. (IEEE ESTEL 2012), Rome, Italy, Oct. 2012, p. 1–6 (2010)Google Scholar
  13. 13.
    Giuliano, R., Mazzenga, F., Neri, A., Vegni, A.M.: Security access protocols in IoT capillary networks. IEEE Internet Things J. 4(3), 645–657 (2017)CrossRefGoogle Scholar
  14. 14.
    Vivado Design Suite UG909 Partial ReconfigurationGoogle Scholar
  15. 15.
    Hochreiter, S., Schmidhuber, J.: Long: Short-Term Memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  16. 16.
    Chang, A.X.M., Culurciello, E.: Hardware accelerators for recurrent neural networks on FPGA. In: 2017 IEEE International Symposium on Circuits and Systems (ISCAS)Google Scholar
  17. 17.
    Krstanovic, S., et al.: Ensembles of recurrent neural networks for robust time series forecasting. In: 2017 International Conference on Innovative Techniques and Applications of AI, CambridgeCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Gian Carlo Cardarilli
    • 1
  • Luca Di Nunzio
    • 1
    Email author
  • Rocco Fazzolari
    • 1
  • Daniele Giardino
    • 1
  • Marco Matta
    • 1
  • Marco Re
    • 1
  • Francesca Silvestri
    • 1
  • Sergio Spanò
    • 1
  1. 1.University of Rome Tor VergataRomeItaly

Personalised recommendations