Advertisement

DIMA Prototype Integrated Circuits

  • Mingu Kang
  • Sujan Gonugondla
  • Naresh R. Shanbhag
Chapter
  • 197 Downloads

Abstract

This chapter describes the design, fabrication and laboratory test results of two DIMA prototype ICs in a 65 nm CMOS process technology. The first IC is a multi-functional DIMA that implements four different machine learning algorithms—the support vector machine (SVM), template matching (TM), k-nearest neighbor (k-NN), and matched filter (MF), thereby demonstrating DIMA’s versatility. The second IC implements the random forest (RF) algorithm which is known be particularly difficult to realize due to its irregular memory accesses. By employing regularized trees and deterministic sub-sampling methods, this IC realizes the RF algorithm via ensemble of decision trees for classification. For both prototypes, design details, including the chip architecture, circuit techniques, and measured results, are provided.

Keywords

Multi-functional in-memory Reconfigurability Support vector machine (SVM) Matched filtering k-Nearest neighbor (k-NN) Template matching Random forest Dot product Machine learning ICs 

References

  1. 1.
    D. Silver, A. Huang, C.J. Maddison, A. Guez, L. Sifre, G. Van Den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot et al., Mastering the game of Go with deep neural networks and tree search. Nature 529(7587), 484–489 (2016)CrossRefGoogle Scholar
  2. 29.
    M. Kang, M.-S. Keel, N.R. Shanbhag, S. Eilert, K. Curewitz, An energy-efficient VLSI architecture for pattern recognition via deep embedding of computation in SRAM, in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2014), pp. 8326–8330Google Scholar
  3. 31.
    J. Zhang, Z. Wang, N. Verma, In-memory computation of a machine-learning classifier in a standard 6T SRAM array. IEEE J. Solid State Circuits 52(4), 915–924 (2017)CrossRefGoogle Scholar
  4. 32.
    M. Kang, S.K. Gonugondla, A. Patil, N.R. Shanbhag, A multi-functional in-memory inference processor using a standard 6T SRAM array. IEEE J. Solid State Circuits 53(2), 642–655 (2018)CrossRefGoogle Scholar
  5. 39.
    C. Sakr, Y. Kim, N. Shanbhag, Analytical guarantees on numerical precision of deep neural networks, in International Conference on Machine Learning (ICML) (2017), pp. 3007–3016Google Scholar
  6. 42.
    M. Kang, S.K. Gonugondla, N.R. Shanbhag, A 19.4 nJ/decision 364K decisions/s in-memory random forest classifier in 6T SRAM array, in IEEE European Solid-State Circuits Conference (ESSCIRC) (2017), pp. 263–266Google Scholar
  7. 44.
    M. Kang, S. Gonugondla, A. Patil, N. Shanbhag, A 481pJ/decision 3.4M decision/s multifunctional deep in-memory inference processor using standard 6T SRAM array. arXiv:1610.07501 (preprint, 2016)Google Scholar
  8. 56.
    Z. Zhou, B. Pain, E.R. Fossum, CMOS active pixel sensor with on-chip successive approximation analog-to-digital converter. IEEE Trans. Electron Devices44(10), 1759–1763 (1997)CrossRefGoogle Scholar
  9. 57.
    F. Arnaud, F. Boeuf, F. Salvetti, D. Lenoble, F. Wacquant, C. Regnier, P. Morin, N. Emonet, E. Denis, J. Oberlin et al., A functional 0.69 μm2 embedded 6T-SRAM bit cell for 65nm CMOS platform, in IEEE Symposium on VLSI Technology (VLSI Technology) (2003), pp. 65–66.Google Scholar
  10. 58.
    H. Kaul, M.A. Anders, S.K. Mathew, G. Chen, S.K. Satpathy, S.K. Hsu, A. Agarwal, R.K. Krishnamurthy, A 21.5 M-query-vectors/s 3.37 nJ/vector reconfigurable k-nearest-neighbor accelerator with adaptive precision in 14nm tri-gate CMOS, in IEEE International Solid-State Circuits Conference (ISSCC) (2016), pp. 260–261Google Scholar
  11. 59.
    S. Gupta, A. Agrawal, K. Gopalakrishnan, P. Narayanan, Deep learning with limited numerical precision, in International Conference on Machine Learning (ICML) (2015), pp. 1737–1746Google Scholar
  12. 60.
    S.C. Chung, Circuits and methods of a self-timed high speed SRAM, Nov. 10, 2015. US Patent 9,183,897Google Scholar
  13. 61.
    E. Karl, Y. Wang, Y.-G. Ng, Z. Guo, F. Hamzaoglu, M. Meterelliyoz, J. Keane, U. Bhattacharya, K. Zhang, K. Mistry et al., A 4.6 GHz 162 Mb SRAM design in 22 nm tri-gate CMOS technology with integrated read and write assist circuitry. IEEE J. Solid State Circuits48(1), 150–158 (2013)Google Scholar
  14. 62.
    Center for biologicaland computational learning (CBCL) at MIT (2000). http://cbcl.mit.edu/software-datasets/index.html
  15. 64.
    Y. LeCun, C. Cortes, MNIST handwritten digit database. AT&T Labs (2010). http://yann.lecun.com/exdb/mnist
  16. 65.
    L. Breiman, Random forests. Mach. Learn. 45(1), 5–32 (2001)CrossRefGoogle Scholar
  17. 66.
    T. Kobayashi, K. Nogami, T. Shirotori, Y. Fujimoto, A current-controlled latch sense amplifier and a static power-saving input buffer for low-power architecture. IEEE J. Solid State Circuits76(5), 863–867 (1993)Google Scholar
  18. 67.
    V.A. Prisacariu, R. Timofte, K. Zimmermann, I. Reid, L. Van Gool, Integrating object detection with 3D tracking towards a better driver assistance system, in IEEE International Conference on Pattern Recognition (ICPR) (2010), pp. 3344–3347Google Scholar
  19. 68.
    B. Van Essen, C. Macaraeg, M. Gokhale, R. Prenger, Accelerating a random forest classifier: multi-core, GP-GPU, or FPGA? in IEEE Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM) (2012), pp. 232–239Google Scholar
  20. 69.
    T.-W. Chen, Y.-C. Su, K.-Y. Huang, Y.-M. Tsai, S.-Y. Chien, L.-G. Chen, Visual vocabulary processor based on binary tree architecture for real-time object recognition in full-HD resolution. IEEE Trans. Very Large Scale Integr. Syst.20(12), 2329–2332 (2012)CrossRefGoogle Scholar
  21. 70.
    K.J. Lee, G. Kim, J. Park, H.-J. Yoo, A vocabulary forest object matching processor with 2.07 M-vector/s throughput and 13.3 nJ/vector per-vector energy for full-HD 60 fps video object recognition. IEEE J. Solid State Circuits50(4), 1059–1069 (2015)Google Scholar
  22. 63.
    Production Crate, Gun shot sounds. http://soundscrate.com/gun-related

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Mingu Kang
    • 1
  • Sujan Gonugondla
    • 2
  • Naresh R. Shanbhag
    • 2
  1. 1.IBM T. J. Watson Research CenterOld TappanUSA
  2. 2.University of Illinois at Urbana-ChampaignUrbanaUSA

Personalised recommendations