Algorithmic-Level Approximate Tensorial SVM Using High-Level Synthesis on FPGA

Approximate Computing Techniques (ACT) are promising solutions towards the achievement of reduced energy, time latency and hardware size for embedded implementations of machine learning algorithms. In this paper, we present the first FPGA implementation of an approximate tensorial Support Vector Machine (SVM) classifier with algorithmic level ACTs using High-Level Synthesis (HLS). A touch modality classification framework was adopted to validate the effectiveness of the proposed implementation. When compared to exact implementation presented in the state-of-the-art, the proposed implementation achieves a reduction in power consumption by up to 49% with a speedup of 3.2×. Moreover, the hardware resources are reduced by 40% while consuming 82% less energy in classifying an input touch with an accuracy loss less than 5%.

[1]  Shereen Afifi,et al.  Hardware Implementations of SVM on FPGA: A State-of-the-Art Review of Current Practice , 2015 .

[2]  Maurizio Valle,et al.  Distributed Sensing and Stimulation Systems for Sense of Touch Restoration in Prosthetics , 2017, 2017 New Generation of CAS (NGCAS).

[3]  Maurizio Valle,et al.  Data Oriented Approximate K-Nearest Neighbor Classifier for Touch Modality Recognition , 2019, 2019 15th Conference on Ph.D Research in Microelectronics and Electronics (PRIME).

[4]  Michele Magno,et al.  An Energy Efficient System for Touch Modality Classification in Electronic Skin Applications , 2019, 2019 IEEE International Symposium on Circuits and Systems (ISCAS).

[5]  Lei Wang,et al.  Reconfigurable Support Vector Machine Classifier with Approximate Computing , 2017, 2017 IEEE Computer Society Annual Symposium on VLSI (ISVLSI).

[6]  Michele Magno,et al.  Energy Efficient System for Tactile Data Decoding Using an Ultra-Low Power Parallel Platform , 2017, 2017 New Generation of CAS (NGCAS).

[7]  Erwan Nogues,et al.  Algorithmic-Level Approximate Computing Applied to Energy Efficient HEVC Decoding , 2019, IEEE Transactions on Emerging Topics in Computing.

[8]  Nikos D. Sidiropoulos,et al.  Tensor Decomposition for Signal Processing and Machine Learning , 2016, IEEE Transactions on Signal Processing.

[9]  M. Sarma,et al.  Implementation of systolic array based SVM classifier using multiplierless kernel , 2014 .

[10]  Antonio J. Plaza,et al.  Approximate Computing of Remotely Sensed Data: SVM Hyperspectral Image Classification as a Case Study , 2016, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[11]  Khaled Benkrid,et al.  Novel dynamic partial reconfiguration implementations of the support vector machine classifier on FPGA , 2016 .

[12]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[13]  Huseyin Seker,et al.  Reconfiguration-based implementation of SVM classifier on FPGA for Classifying Microarray data , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[14]  Sparsh Mittal,et al.  A Survey of Techniques for Approximate Computing , 2016, ACM Comput. Surv..

[15]  Richard P. Brent,et al.  Efficient one-sided Jacobi algorithms for singular value decomposition and the symmetric eigenproblem , 1995, Proceedings 1st International Conference on Algorithms and Architectures for Parallel Processing.

[16]  Maurizio Valle,et al.  Computational Intelligence Techniques for Tactile Sensing Systems , 2014, Sensors.

[17]  Maurizio Valle,et al.  Approximate Computing Methods for Embedded Machine Learning , 2018, 2018 25th IEEE International Conference on Electronics, Circuits and Systems (ICECS).

[18]  Himansu Sekhar Behera,et al.  A Comprehensive Survey on Support Vector Machine in Data Mining Tasks: Applications & Challenges , 2015 .

[19]  Johan A. K. Suykens,et al.  A kernel-based framework to tensorial data analysis , 2011, Neural Networks.

[20]  Ladislav A. Novak,et al.  Reconfigurable Hardware for Machine Learning Applications , 2015, J. Circuits Syst. Comput..

[21]  Zhongfeng Wang,et al.  Energy efficient SVM classifier using approximate computing , 2017, 2017 IEEE 12th International Conference on ASIC (ASICON).

[22]  Maurizio Valle,et al.  Algorithmic Level Approximate Computing for Machine Learning Classifiers , 2019, 2019 26th IEEE International Conference on Electronics, Circuits and Systems (ICECS).

[23]  Maurizio Valle,et al.  Real-Time Embedded Machine Learning for Tensorial Tactile Data Processing , 2018, IEEE Transactions on Circuits and Systems I: Regular Papers.