Abstract
This paper presents a new approach for support vector filtering to accelerate the training process of support vector machines (SVMs). It is based on the Sets-of-Real-Numbers (SORN) number format, which provides low complex and ultra fast computing. SORNs are an interval based binary number format, showing promising results for complex arithmetic operations, e.g. multiplication or fused multiply-add. To apply SORNs to high dimensional vector arithmetic, a combination of SORN arithmetic and fixed-point adder trees is used. This Hybrid SORN approach combines the advantages of SORNs, concerning reduction of computational costs and time, and fixed point adders in terms of precision. A Hybrid SORN support vector filtering architecture is implemented on an FPGA board with Zynq 7000 XC7Z100 SoC and evaluated for the MNIST dataset. It can be considered as hardware accelerator, reducing the training time by factor 1.38 for one-versus-rest and 2.65 for one-versus-one SVM implementation.
The authors acknowledge the financial support by the Federal Ministry of Education and Research of Germany in the project “Open6GHub” (grant number: 16KISK016).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hülsmeier, N., Bärthel, M., Karsthof, L., Rust, J., Paul, S.: Hybrid SORN Implementation of k-nearest neighbor algorithm on FPGA. In: 2022 20th IEEE Interregional NEWCAS Conference (NEWCAS), Quebec, Canada (2022)
Cortes, C., Vapnik, V.: Support vector networks. Mach. Learn. 20, 273–297 (1995)
Graf, H.P., Cosatto, E., Bottou, L., Dourdanovic, I., Vapnik, V.: Parallel support vector machines: the cascade SVM. Proc. Adv. Neural Inf. Process. Syst. 521–528 (2004)
Gustafson, J.: A radical approach to computation with real numbers. Supercomput. Front. Innov. 3(2) (2016)
Hülsmeier, N., Bärthel, M., Rust, J., Paul, S.: SORN-based cascade support vector machine. In: 28th European Signal Processing Conference (EUSIPCO), Amsterdam, Netherlands (2021)
Bärthel, M., Seidel, P., Rust, J., Paul, S.: SORN arithmetic for MIMO symbol detection - exploration of the type-2 Unum format. In: 2019 17th IEEE International New Circuits and Systems Conference (NEWCAS), Munich, Germany (2019)
Bärthel, M., Rust, J., Paul, S.: Application-specific analysis of different SORN datatypes for Unum type-2-based arithmetic. In: 2020 IEEE International Symposium on Circuits and Systems (ISCAS), Sevilla, Spain (2020)
Bärthel, M., Hülsmeier, N., Rust, J., Paul, S.: On the implementation of edge detection algorithms with SORN arithmetic. In: Gustafson, J., Dimitrov, V. (eds.) Next Generation Arithmetic. CoNGA 2022. LNCS, vol. 13253, pp. 1–13. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-09779-9_1
Bärthel, M., Knobbe, S., Rust, J., Paul, S.: Hardware implementation of a latency-reduced sphere decoder with SORN preprocessing. IEEE Access 9, 91387–91401 (2021)
AXI Dma v7.1 LogiCORE IP Product Guide Vivado Design Suite, Xilinx, July 2019. https://www.xilinx.com/support/documentation/ip_documentation/axi_dma/v7_1/pg021_axi_dma.pdf. Accessed 27 Nov 2022
LeCun, Y., Cortes, C., Burges, C.: MNIST handwritten digit database, ATT Labs, vol. 2 (2010). http://yann.lecun.com/exdb/mnist. Accessed 10 Mar 2021
Wang, Q., Li, P., Kim, Y.: A parallel digital VLSI architecture for integrated support vector machine training and classification. In: IEEE Integration on Very Large Scale Integration (VLSI) Systems, vol. 23, no. 8 (2015)
Afifi, S., GholamHosseini, H., Sinha, R.: FPGA implementations of SVM classifiers: a review. SN Comput. Sci. 1(133) (2020)
Romanov, A., et al.: Analysis of posit and Bfloat arithmetic of real numbers for machine learning. IEEE Access 9, 82318–82324 (2021)
Pedregosa, F., Varoquaux, G., Gramfort, A., Prettenhofer, P.: Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Hülsmeier, N., Bärthel, M., Rust, J., Paul, S. (2023). Hybrid SORN Hardware Accelerator for Support Vector Machines. In: Gustafson, J., Leong, S.H., Michalewicz, M. (eds) Next Generation Arithmetic. CoNGA 2023. Lecture Notes in Computer Science, vol 13851. Springer, Cham. https://doi.org/10.1007/978-3-031-32180-1_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-32180-1_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-32179-5
Online ISBN: 978-3-031-32180-1
eBook Packages: Computer ScienceComputer Science (R0)