Cited By
View all- Klachko MMahmoodi MStrukov D(2019)Improving Noise Tolerance of Mixed-Signal Neural Networks2019 International Joint Conference on Neural Networks (IJCNN)10.1109/IJCNN.2019.8851966(1-8)Online publication date: Jul-2019
Mixed-Signal POp/J Computing with Nonvolatile Memories
Page 513
Abstract
The present-day revolution in deep learning was triggered not by any significant algorithm breakthrough, but by the use of more powerful GPU hardware [1]. Though this revolution has stimulated the development of even more powerful dedicated digital systems [2, 3], their speed and energy efficiency are still insufficient for ultrafast pattern classification and more ambitious cognitive tasks. The main reason is that the use of digital operations for the implementation of neuromorphic networks, with their high redundancy and noise/variability tolerance, is inherently unnatural. On the other hand, the network performance may be dramatically improved using mixed-signal integrated circuits, where the key inference-stage operation, the vector-by-matrix multiplication, is implemented on the physical level by utilization of the fundamental Ohm and Kirchhoff laws [4-6].
References
[1]
Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. 2017. ImageNet classification with deep convolutional neural networks. Commun. ACM 60, 6 (May 2017), 84--90.
[2]
Y. H. Chen, T. Krishna, J. Emer, and V. Sze. 2016. Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks. IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, 262--263.
[3]
B. Moons, R. Uytterhoeven, W. Dehaene, and M. Verhelst, 2017. Envision: A 0.26-to-10 TOps/W subword-parallel dynamic-voltage-accuracy-frequency-scalable convolutional neural network processor in 28 nm FDSOI. IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, 246--247.
[4]
C. A. Mead. 1989. Analog VLSI and Neural Systems. Addison Wesley.
[5]
D. Strukov, F. Merrikh-Bayat, M. Prezioso, X. Guo, B. Hoskins and K. Likharev. 2015. Memory technologies for neural networks" IEEE International Memory Workshop (IMW), Monterey, CA, 1--4.
[6]
L. Ceze et al. 2016. Nanoelectronic neurocomputing: Status and prospects. 2016 74th Annual Device Research Conference (DRC), Newark, DE, 1--2.
[7]
F. M. Bayat, X. Guo, H. A. Om'mani, N. Do, K. K. Likharev and D. B. Strukov. 2015. Redesigning commercial floating-gate memory for analog computing applications. IEEE International Symposium on Circuits and Systems (ISCAS), Lisbon, 1921--1924.
[8]
F. M. Bayat, X. Guo, M. Klachko, N. Do, K. Likharev and D. Strukov. 2016. Model-based high-precision tuning of NOR flash memory cells for analog computing applications. 74th Annual Device Research Conference (DRC), Newark, DE, 1--2.
[9]
X. Guo et al. 2017. Temperature-insensitive analog vector-by-matrix multiplier based on 55-nm NOR flash memory cells. IEEE Custom Integrated Circuits Conference (CICC), Austin, TX, 1--4.
[10]
F. Merrikh Bayat et al. 2018. High-performance mixed-signal neurocomputing with nanoscale floating-gate memory cells. IEEE Transactions on Neural Networks and Learning Systems.
[11]
X. Guo et al. 2017. Fast, energy-efficient, robust, and reproducible mixed-signal neuromorphic classifier based on embedded NOR flash memory technology. IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, 6.5.1--6.5.4.
[12]
S. K. Esser, R. Appuswamy, P. Merolla, J. V. Arthur, and D. S. Modha, 2015. Backpropagation for energy-efficient neuromorphic computing. Proc. NIPS'15, Montreal, Canada, 1117--1125.
[13]
M. Prezioso et al. 2015. Training and operation of an integrated neuromorphic network based on metal-oxide memristors", Nature, 521, (2015) pp.61--64.
[14]
M. Prezioso et al. 2015. Modeling and implementation of firing rate neuromorphic-network classifiers with bilayer Pt/Al2O3/TiO2-x/Pt memristors. IEEE International Electron Devices Meeting (IEDM), Washington, DC, 17.4.1--17.4.4.
[15]
F. Merrikh Bayat, M. Prezioso, B. Chakrabarti, I. Kataeva, and D. Strukov. 2017. Memristor-based perceptron classifier: Increasing complexity and coping with imperfect hardware. IEEE/ACM International Conference on Computer-Aided Design (ICCAD), Irvine, CA, 549--554.
[16]
F. Merrikh Bayat, M. Prezioso, B. Chakrabarti, I. Kataeva, and D. B. Strukov. 2017. Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits. ArXiv:1611.04465, submitted to Nature Comm., Nov. 2017.
[17]
G. C. Adam et al. 2017. 3-D memristor crossbars for analog and neuromorphic computing applications", IEEE Transactions on Electron Devices, vol. 64 (2017), 312--318.
[18]
B. Chakrabarti et al. 2017. A multiply-add engine with monolithically integrated 3D memristor crossbar/CMOS hybrid circuit. Nature Scientific Reports, vol. 7 (2017). art. 42429.
[19]
M.R. Mahmoodi and D.B. Strukov. 2018. An ultra-low energy internally analog, externally digital vector-matrix multiplier circuit based on NOR flash memory technology. Proc. DAC'18, San Francisco, C,A 1--6.
[20]
M. Bavandpour, M. R. Mahmoodi, and D. B. Strukov. 2017. Energy-efficient time-domain vector-by-matrix multiplier for neurocomputing and beyond. ArXiv:1711.10673.
Index Terms
- Mixed-Signal POp/J Computing with Nonvolatile Memories
Recommendations
The zen of nonvolatile memories
DAC '06: Proceedings of the 43rd annual Design Automation ConferenceSilicon technology based nonvolatile memories (NVM) have achieved widespread adoption for code and data storage applications. In the last 30 years, the traditional oating gate bitcell has been scaled following Moore's law, but recently scaling limits ...
Accelerated design of analog, mixed-signal circuits in Titan
ISPD '09: Proceedings of the 2009 international symposium on Physical designAnalog circuits are designed in fundamentally different ways than their counterparts in the digital domain. However, two major trends have made analog and digital circuits mix and co-exist. First, high-performance analog circuits use digital parts and ...
Automatic test vector generation for mixed-signal circuits
EDTC '95: Proceedings of the 1995 European conference on Design and TestMixed circuit testing is known to be a very difficult task. This is due to the difficulty of: testing the analog part of the circuit, controlling the digital signal from the analog outputs, observing the analog outputs in the digital circuit, ...
Comments
Information & Contributors
Information
Published In
May 2018
533 pages
ISBN:9781450357241
DOI:10.1145/3194554
- General Chair:
- Deming Chen,
- Program Chairs:
- Houman Homayoun,
- Baris Taskin
Copyright © 2018 Owner/Author.
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author.
Sponsors
- SIGDA: ACM Special Interest Group on Design Automation
- IEEE CEDA
- IEEE CASS
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 30 May 2018
Check for updates
Author Tags
Qualifiers
- Abstract
Conference
GLSVLSI '18
Sponsor:
Acceptance Rates
GLSVLSI '18 Paper Acceptance Rate 48 of 197 submissions, 24%;
Overall Acceptance Rate 312 of 1,156 submissions, 27%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 243Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 13 Jan 2025
Other Metrics
Citations
Cited By
View all- Klachko MMahmoodi MStrukov D(2019)Improving Noise Tolerance of Mixed-Signal Neural Networks2019 International Joint Conference on Neural Networks (IJCNN)10.1109/IJCNN.2019.8851966(1-8)Online publication date: Jul-2019
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in