nach oben

Erschienen in:

2024 | OriginalPaper | Buchkapitel

2. Basics and Research Status of Neural Network Processors

verfasst von : Jinshan Yue

Erschienen in: High Energy Efficiency Neural Network Processor with Combined Digital and Computing-in-Memory Architecture

Verlag: Springer Nature Singapore

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

Das Kapitel beginnt mit einem historischen Überblick über neuronale Netzwerkalgorithmen, der ihre Entwicklung von frühen Modellen wie dem M-P-Modell und der einschichtigen Wahrnehmung bis hin zu modernen Weiterentwicklungen wie Deep Learning und Transformatoren verfolgt. Anschließend vertieft er sich in die Grundlagen neuronaler Netzwerkprozessoren und diskutiert Schlüsselkennzahlen wie Leistung, Leistung und Energieeffizienz. Der Forschungsstand digitaler schaltungsbasierter neuronaler Netzwerkprozessoren wird untersucht, wobei der Schwerpunkt auf Optimierungstechniken wie Datenwiederverwendung, Low-Bit-Quantisierung und Komprimierung von Netzwerkmodellen liegt. Das Kapitel behandelt auch Computing-in-Memory neuronale Netzwerkprozessoren und diskutiert ihre Prinzipien, Bauteile, Schaltkreise und Makroebenen-Designs. Er hebt das Potenzial von CIM für eine höhere Energieeffizienz hervor, verweist aber auch auf die Herausforderungen und unerforschten Forschungsräume. Das Kapitel schließt mit einer Zusammenfassung der bestehenden Herausforderungen und der Notwendigkeit weiterer Forschung sowohl in digitalen Schaltkreisen als auch in CIM-Prozessoren für neuronale Netzwerke.

KI-Generiert

Diese Zusammenfassung des Fachinhalts wurde mit Hilfe von KI generiert.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 390 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Maschinenbau + Werkstoffe

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Vorheriges Kapitel Introduction

Nächstes Kapitel Energy-Efficient NN Processor by Optimizing Data Reuse for Specific Convolutional Kernels

James W (1990) The principles of psychology. In: Great books of the Western world, 53

McClulloch WS, Pitts W (1943) A logical calculus of the ideas immanent in neurons activity. Bull Math Biophys 5(115–133):10

Rosenblatt F (1958) The perceptron: a probabilistic model for information storage and organization in the brain. Psychol Rev 65(6):386CrossRef

Minsky M, Papert S (1969) Perceptron: an introduction to computational geometry. The MIT Press Cambrid Expanded Edn 19(88):2

Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323(6088):533–536

LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324CrossRef

LeCun Y, Cortes C, Burges CJC (1994) The MNIST database of handwritten digits

Hinton GE, Osindero S, Teh Y-W (2006) A fast learning algorithm for deep belief nets. Neural Comp 18(7):1527–1554

Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L (2009) ImageNet: A large-scale hierarchical image database. In: 2009 IEEE conference on computer vision and pattern recognition. IEEE, pp 248–255

10.

Sermanet P, Eigen D, Zhang X, Mathieu M, Fergus R, LeCun Y (2013) Overfeat: integrated recognition, localization and detection using convolutional networks. arXiv:1312.6229

11.

Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2014) Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1–9

12.

Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556

13.

He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778

14.

Huang G, Liu Z, Maaten LVD, Weinberger KQ (2017) Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4700–4708

15.

Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 7132–7141

16.

Hochreiter S, Schmidhuber J, Elvezia C (1997) Long short-term memory. Neural Comput 9(8):1735–1780

17.

Ali Z (2001) Fundamentals of neural networks. Intell Control Syst Using Soft Comput Methodol 1:1–5

18.

Sak H, Senior AW, Beaufays F (2014) Long short-term memory recurrent neural network architectures for large scale acoustic modeling. Int J Speech Technol

19.

Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. In: Proceedings of the 31st international conference on neural information processing systems, pp 6000–6010

20.

Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst 27

21.

Silver D, Huang A, Maddison CJ, Guez A, Sifre L, Driessche GVD, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M et al (2016) Mastering the game of go with deep neural networks and tree search. Nature 529(7587):484–489

22.

Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105

23.

NVIDIA A100 tensor core GPU (2021)

24.

Howard AG, Zhu M, Chen B, Kalenichenko D, Wang W, Weyand T, Andreetto M, Adam H (2017) Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861

25.

Zhang X, Zhou X, Lin M, Sun J (2018) Shufflenet: an extremely efficient convolutional neural network for mobile devices. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 6848–6856

26.

Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Math Control Signals Syst 2(4):303–314MathSciNetCrossRef

27.

Nair V, Hinton GE (2010) Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th international conference on international conference on machine learning, pp 807–814

28.

Jarrett K, Kavukcuoglu K, Ranzato MA, LeCun Y (2009) What is the best multi-stage architecture for object recognition? In: 2009 IEEE 12th international conference on computer vision. IEEE, pp 2146–2153

29.

Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: International conference on machine learning. PMLR, pp 448–456

30.

LiKamWa R, Hou Y, Gao J, Polansky M, Zhong L (2016) Redeye: analog convnet image sensor architecture for continuous mobile vision. ACM SIGARCH Comput Archit News 44(3):255–266CrossRef

31.

Li Q, Zhu H, Qiao F, Liu X, Wei Q, Yang H (2018) Energy-efficient MFCC extraction architecture in mixed-signal domain for automatic speech recognition. In: 2018 IEEE/ACM international symposium on nanoscale architectures (NANOARCH). IEEE, pp 1–3

32.

Chen Y-H, Krishna T, Emer J, Sze V (2016) 14.5 Eyeriss: an energy-efficient reconfigurable accelerator for deep convolutional neural networks. In: 2016 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 262–264

33.

Moons B, Uytterhoeven R, Dehaene W, Verhelst M (2017) 14.5 ENVISION: a 0.26-to-10 TOPS/W subword-parallel dynamic-voltage-accuracy-frequency-scalable convolutional neural network processor in 28 nm FDSOI. In: 2017 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 246, 247

34.

Jouppi NP, Young C, Patil N, Patterson D, Agrawal G, Bajwa R, Bates S, Bhatia S, Boden N, Borchers A et al (2017) In-datacenter performance analysis of a tensor processing unit. In: Proceedings of the 44th annual international symposium on computer architecture, pp 1–12

35.

Bang S, Wang J, Li Z, Cao G, Sylvester D (2017) 14.7 A 288 \(\upmu \)w programmable deep-learning processor with 270kb on-chip weight storage using non-uniform memory hierarchy for mobile intelligence. In: 2017 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 250, 251

36.

Alwani M, Chen H, Ferdman M, Milder P (2016) Fused-layer CNN accelerators. In: 2016 49th annual IEEE/ACM international symposium on microarchitecture (MICRO). IEEE, pp 1–12

37.

Lee J, Kim C, Kang S, Shin D, Kim S, Yoo H-J (2018) UNPU: a 50.6 TOPS/W unified deep neural network accelerator with 1b-to-16b fully-variable weight bit-precision. In: 2018 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 218–220

38.

Horowitz M (2014) 1.1 computing’s energy problem (and what we can do about it). In: 2014 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 10–14

39.

Guo K, Li W, Zhong K, Zhu Z, Zeng S, Song H, Yuan X, Debacker P, Verhelst M, Wang Y (2020) Neural network accelerator comparison

40.

Yin S, Peng O, Yang J, Lu T, Li X, Liu L, Wei S (2018) An ultra-high energy-efficient reconfigurable processor for deep neural networks with binary/ternary weights in 28 nm CMOS. In: 2018 IEEE symposium on VLSI circuits. IEEE, pp 37–38

41.

Ueyoshi K, Ando K, Hirose K, Takamaeda-Yamazaki S, Kadomoto J, Miyata T, Hamada M, Kuroda T, Motomura M (2018) QUEST: A 7.49 TOPS multi-purpose log-quantized DNN inference engine stacked on 96 MB 3D SRAM using inductive-coupling technology in 40 nm CMOS. In: 2018 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 216–218

42.

Yuan Z, Yue J, Yang H, Wang Z, Li J, Yang Y, Guo Q, Li X, Chang M-F, Yang H et al (2018) STICKER: a 0.41–62.1 TOPS/W 8 bit neural network processor with multi-sparsity compatible convolution arrays and online tuning acceleration for fully connected layers. In: 2018 IEEE symposium on VLSI circuits. IEEE, pp 33–34

43.

Han S, Pool J, Tran J, Dally W (2015) Learning both weights and connections for efficient neural network. In: Advances in neural information processing systems, pp 1135–1143

44.

Wen W, Wu C, Wang Y, Chen Y, Li H (2016) Learning structured sparsity in deep neural networks. Adv Neural Inf Process Syst 29:2074–2082

45.

Mao H, Han S, Pool J, Li W, Liu X, Wang Y, Dally WJ (2017) Exploring the granularity of sparsity in convolutional neural networks, pp 13–20

46.

Zhang T, Ye S, Zhang K, Tang, Wen W, Fardad M, Wang Y (2018) A systematic DNN weight pruning framework using alternating direction method of multipliers. In: Proceedings of the European conference on computer vision (ECCV), pp 184–199

47.

Han S, Liu X, Mao H, Pu J, Pedram A, Horowitz MA, Dally WJ (2016) EIE: efficient inference engine on compressed deep neural network. In: 2016 ACM/IEEE 43rd annual international symposium on computer architecture (ISCA). IEEE, pp 243–254

48.

Parashar A, Rhu M, Mukkara A, Puglielli A, Venkatesan R, Khailany B, Emer J, Keckler SW, Dally WJ (2017) SCNN: an accelerator for compressed-sparse convolutional neural networks. In: 2017 ACM/IEEE 44th annual international symposium on computer architecture (ISCA). IEEE, pp 27–40

49.

Whatmough PN, Lee SK, Lee H, Rama S, Brooks D, Wei G-Y (2017) 14.3 A 28 nm SoC with a 1.2 GHz 568 nJ/prediction sparse deep-neural-network engine with \(>\)0.1 timing error rate tolerance for IoT applications. In: 2017 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 242–243

50.

Mathieu M, Henaff M, LeCun Y (2013) Fast training of convolutional networks through FFTs. arXiv:1312.5851

51.

Vasilache N, Johnson J, Mathieu M, Chintala S, Piantino S, LeCun Y (2014) Fast convolutional nets with FBFFT: a GPU performance evaluation. arXiv:1412.7580

52.

Biswas A, Chandrakasan AP (2018) Conv-RAM: an energy-efficient SRAM with embedded convolution computation for low-power CNN-based machine learning applications. In: 2018 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 488–490

53.

Chi P, Li S, Xu C, Tao Z, Zhao J, Liu Y, Yu W, Xie Y (2016) PRIME: a novel processing-in-memory architecture for neural network computation in ReRAM-based main memory. ACM SIGARCH Comput Archit News 44(3):27–39

54.

Shafiee A, Nag A, Muralimanohar N, Balasubramonian R, Strachan JP, Hu M, Stanley Williams R, Srikumar V (2016) ISAAC: a convolutional neural network accelerator with in-situ analog arithmetic in crossbars. ACM SIGARCH Comput Archit News 44(3):14–26

55.

Song L, Qian X, Li H, Chen Y (2017) Pipelayer: a pipelined ReRAM-based accelerator for deep learning. In: 2017 IEEE international symposium on high performance computer architecture (HPCA). IEEE, pp 541–552

56.

Yue J, Liu Y, Su F, Li S, Yuan Z, Wang Z, Sun W, Li X, Yang H (2019) AERIS: area/energy-efficient 1T2R ReRAM based processing-in-memory neural network system-on-a-chip. In: Proceedings of the 24th Asia and South Pacific design automation conference, pp 146–151

57.

Khwa W-S, Chen J-J, Li J-F, Si X, Yang E-Y, Sun X, Liu R, Chen P-Y, Li Q, Yu S et al (2018) A 65 nm 4 kB algorithm-dependent computing-in-memory SRAM unit-macro with 2.3 ns and 55.8 TOPS/W fully parallel product-sum operation for binary DNN edge processors. In: 2018 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 496–498

58.

Si X, Chen J-J, Tu Y-N, Huang W-H, Wang J-H, Chiu Y-C, Wei W-C, Wu S-Y, Sun X, Liu R et al (2019) 24.5 A twin-8T SRAM computation-in-memory macro for multiple-bit CNN-based machine learning. In: 2019 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 396–398

59.

Li S, Niu D, Malladi KT, Zheng H, Brennan B, Xie Y (2017) DRISA: a DRAM-based reconfigurable in-situ accelerator. In: 2017 50th annual IEEE/ACM international symposium on microarchitecture (MICRO). IEEE, pp 288–301

60.

Seshadri V, Lee D, Mullins T, Hassan H, Boroumand A, Kim J, Kozuch MA, Mutlu O, Gibbons PB, Mowry TC (2017) Ambit: in-memory accelerator for bulk bitwise operations using commodity DRAM technology. In: 2017 50th annual IEEE/ACM international symposium on microarchitecture (MICRO). IEEE, pp 273–287

61.

Guo X, Bayat FM, Prezioso M, Chen Y, Nguyen B, Do N, Strukov DB (2017) Temperature-insensitive analog vector-by-matrix multiplier based on 55 nm NOR flash memory cells. In: 2017 IEEE custom integrated circuits conference (CICC). IEEE, pp 1–4

62.

Guo X, Bayat FM, Bavandpour M, Klachko M, Mahmoodi MR, Prezioso M, Likharev KK, Strukov DB (2017) Fast, energy-efficient, robust, and reproducible mixed-signal neuromorphic classifier based on embedded NOR flash memory technology. In: 2017 IEEE international electron devices meeting (IEDM). IEEE, pp 6.5.1–6.5.4

63.

Chen W-H, Li K-X, Lin W-Y, Hsu K-H, Li P-Y, Yang C-H, Xue C-X, Yang E-Y, Chen Y-K, Chang Y-S et al (2018) A 65 nm 1 MB nonvolatile computing-in-memory ReRAM macro with sub-16 ns multiply-and-accumulate for binary DNN AI edge processors. In: 2018 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 494–496

64.

Liu Q, Gao B, Yao P, Wu D, Chen J, Pang Y, Zhang W, Liao Y, Xue C-X, Chen W-H et al (2020) 33.2 A fully integrated analog ReRAM based 78.4 TOPS/W compute-in-memory chip with fully parallel MAC computing. In: 2020 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 500–502

65.

Keshavarzi A, Ni K, Van Den H Wilbert, Datta S, Raychowdhury A (2020) Ferroelectronics for edge intelligence. IEEE Micro 40(6):33–48CrossRef

66.

Khan AI, Keshavarzi A, Datta S (2020) The future of ferroelectric field-effect transistor technology. Nat Electron 3(10):588–597

67.

Le Gallo M, Sebastian A, Mathis R, Manica M, Giefers H, Tuma T, Bekas C, Curioni A, Eleftheriou E (2018) Mixed-precision in-memory computing. Nat Electron 1(4):246–253CrossRef

68.

Shi Y, Oh S, Huang Z, Lu X, Kang SH, Kuzum D (2020) Performance prospects of deeply scaled spin-transfer torque magnetic random-access memory for in-memory computing. IEEE Electron Dev Lett 41(7):1126–1129

69.

Yang J, Kong Y, Wang Z, Liu Y, Wang B, Yin S, Shi L (2019) 24.4 sandwich-RAM: an energy-efficient in-memory BWN architecture with pulse-width modulation. In: 2019 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 394–396

70.

Si X, Tu Y-N, Huanq W-H, Su J-W, Lu P-J, Wang J-H, Liu T-W, Wu S-Y, Liu R, Chou Y-C et al (2020) 15.5 A 28 nm 64 kB 6T SRAM computing-in-memory macro with 8b MAC operation for AI edge chips. In: 2020 IEEE international solid-state circuits conference (ISSCC). IEEE, pp 246–248

71.

Bankman D, Yang L, Moons B, Verhelst M, Murmann B (2018) An always-on 3.8 \(\upmu \)J/86% CIFAR-10 mixed-signal binary CNN processor with all memory on chip in 28-nm CMOS. IEEE J Sol-State Circ 54(1):158–172

72.

Valavi H, Ramadge PJ, Nestler E, Verma N (2019) A 64-tile 2.4-MB in-memory-computing CNN accelerator employing charge-domain compute. IEEE J Sol-State Circ 54(6):1789–1799

73.

Jia H, Ozatay M, Tang Y, Valavi H, Pathak R, Lee J, Verma N (2021) A programmable neural-network inference accelerator based on scalable in-memory computing. In: 2021 IEEE international solid-state circuits conference (ISSCC), vol 64. IEEE, pp 236–238

74.

Guo R, Liu Y, Zheng S, Wu S-Y, Ouyang P, Khwa W-S, Chen X, Chen J-J, Li X, Liu L et al (2019) A 5.1 pJ/neuron 127.3 \(\upmu \)s/inference RNN-based speech recognition processor using 16 computing-in-memory SRAM macros in 65 nm CMOS. In: 2019 symposium on VLSI circuits. IEEE, pp C120–C121

Titel: Basics and Research Status of Neural Network Processors
verfasst von: Jinshan Yue
Verlag: Springer Nature Singapore
Buch: High Energy Efficiency Neural Network Processor with Combined Digital and Computing-in-Memory Architecture
Print ISBN: 978-981-9734-76-4

Electronic ISBN: 978-981-9734-77-1

Copyright-Jahr: 2024
DOI: https://doi.org/10.1007/978-981-97-3477-1_2