[논문]Exploiting defective RRAM array as synapses of HTM spatial pooler with boost-factor adjustment scheme for defect-tolerant neuromorphic systems

Woo, Jiyong; Van Nguyen, Tien; Kim, Jeong Hun; Im, Jong-Pil; Im, Solyee; Kim, Yeriaron; Min, Kyeong-Sik; Moon, Seung Eon

doi:10.1038/s41598-020-68547-5

[해외논문] Exploiting defective RRAM array as synapses of HTM spatial pooler with boost-factor adjustment scheme for defect-tolerant neuromorphic systems 원문보기

Scientific reports, v.10, 2020년, pp.11703 -

Woo, Jiyong (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea) , Van Nguyen, Tien (School of Electrical Engineering, Kookmin University, Seoul, 02707 South Korea) , Kim, Jeong Hun (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea) , Im, Jong-Pil (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea) , Im, Solyee (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea) , Kim, Yeriaron (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea) , Min, Kyeong-Sik (School of Electrical Engineering, Kookmin University, Seoul, 02707 South Korea) , Moon, Seung Eon (ICT Creative Research Laboratory, Electronics and Telecommunications Research Institute, Daejeon, 34129 South Korea)

Abstract ▼ AI-Helper

A crossbar array architecture employing resistive switching memory (RRAM) as a synaptic element accelerates vector–matrix multiplication in a parallel fashion, enabling energy-efficient pattern recognition. To implement the function of the synapse in the RRAM, multilevel resistance states are required. More importantly, a large on/off ratio of the RRAM should be preferentially obtained to ensure a reasonable margin between each state taking into account the inevitable variability caused by the inherent switching mechanism. The on/off ratio is basically adjusted in two ways by modulating measurement conditions such as compliance current or voltage pulses modulation. The latter technique is not only more suitable for practical systems, but also can achieve multiple states in low current range. However, at the expense of applying a high negative voltage aimed at enlarging the on/off ratio, a breakdown of the RRAM occurs unexpectedly. This stuck-at-short fault of the RRAM adversely affects the recognition process based on reading and judging each column current changed by the multiplication of the input voltage and resistance of the RRAM in the array, degrading the accuracy. To address this challenge, we introduce a boost-factor adjustment technique as a fault-tolerant scheme based on simple circuitry that eliminates the additional process to identify specific locations of the failed RRAMs in the array. Spectre circuit simulation is performed to verify the effect of the scheme on Modified National Institute of Standards and Technology dataset using convolutional neural networks in non-ideal crossbar arrays, where experimentally observed imperfective RRAMs are configured. Our results show that the recognition accuracy can be maintained similar to the ideal case because the interruption of the failure is suppressed by the scheme.

참고문헌 (49)

1. Merolla P A million spiking-neuron integrated circuit with a scalable communication network and interface Science 2014 345 6197 668 673 10.1126/science.1254642 25104385

상세보기
2. LuCun Y Bengio Y Hinton G Deep learning Nature 2015 521 7553 436 444 10.1038/nature14539 26017442

상세보기
3. Hu, M. et al., Dot-product engine for neuromorphic computing: Programming 1T1M crossbar to accelerate matrix-vector multiplication. in ACM/EDAC/IEEE Design Automation Conference (DAC) (2016).
4. Vanhoucke, V., Senior, A. & Mao, M. Improving the speed of neural networks on CPUs. in Deep learning and unsupervised feature learning workshop (NIPS) . 1？8 (2011).
5. Krizhevsky, A., Sutskever, I. & Hinton, G. ImageNet classification with deep convolutional neural networks. in Advances in Neural Information Processing Systems (NIPS) . 1？9 (2012).
6. Ielmini D Wong H-SP In-memory computing with resistive switching devices Nat. Eelectron. 2018 1 333 343 10.1038/s41928-018-0092-2

상세보기
7. Xia Q Memristor-CMOS hybrid integrated circuits for reconfigurable logic Nano Lett. 2009 9 10 3640 3645 10.1021/nl901874j 19722537

상세보기
8. Yu S Neuro-inspired computing with emerging nonvolatile memorys Proc. IEEE 2018 106 2 260 285 10.1109/JPROC.2018.2790840

상세보기
9. Ambrogio S Equivalent-accuracy accelerated neural-network training using analogue memory Nature 2018 558 60 67 10.1038/s41586-018-0180-5 29875487

상세보기
10. Srinivasan G Sengupta A Roy K Magnetic tunnel junction based long-term short-term stochastic synapse for a spiking neural network with on-chip STDP learning Sci. Rep. 2016 6 29545 10.1038/srep29545 27405788

상세보기
11. Boyn S Learning through ferroelectric domain dynamics in solid-state synapses Nat. Commun. 2017 8 14736 10.1038/ncomms14736 28368007

상세보기
12. Fuller E Li-ion synaptic transistor for low power analog computing Adv. Mater. 2017 29 4 1604310 10.1002/adma.201604310

상세보기
13. Yang J Strukov D Stewart D Memristive devices for computing Nat. Nanotechnol. 2013 8 1 13 10.1038/nnano.2012.240 23269430

상세보기
14. Woo J Improved synaptic behavior under identical pulses using AlO x /HfO 2 bilayer RRAM array for neuromorphic systems IEEE Electron Device Lett. 2016 37 8 994 997 10.1109/LED.2016.2582859

상세보기
15. Pedretti G Memristive neural network for on-line learning and tracking with brain-inspired spike timing dependent plasticity Sci. Rep. 2017 7 5288 10.1038/s41598-017-05480-0 28706303

상세보기
16. Chakrabarti B A multiply-add engine with monolithically integrated 3D memristor crossbar/CMOS hybrid circuit Sci. Rep. 2017 7 42429 10.1038/srep42429 28195239

상세보기
17. Woo J Yu S Resistive memory-based analog synapse: the pursuit for linear and symmetric weight update IEEE Nanatechnol. Mag. 2018 12 3 36 44 10.1109/MNANO.2018.2844902

상세보기
18. Prakash A Demonstration of low power 3-bit multilevel cell characteristics in a TaO x -based RRAM by stack engineering IEEE Electron Device Lett. 2015 36 1 32 34 10.1109/LED.2014.2375200

상세보기
19. Stathopoulos S Multibit memory operation of metal-oxide bi-layer memristors Sci. Rep. 2017 7 17532 10.1038/s41598-017-17785-1 29235524

상세보기
20. Xia, Q. et al., Memristive crossbar arrays for brain-inspired computing. nature materials, 18, pp. 309？323 (2019).
21. Woo J Impact of selector devices in analog RRAM-based crossbar arrays for inference and training of neuromorphic system IEEE Trans. Very Large Scale Integr. Syst. 2019 27 9 2205 2212 10.1109/TVLSI.2019.2917764

상세보기
22. Chen, P. & Yu, S. Stuck-at-fault tolerant schemes for memristor crossbar array-based neural networks. in IEEE International Reliability Physics Symposium . 5C.4.1？5C.4.4 (2018).
23. Wu, H. et al., Reliability perspective on neuromorphic computing based on analog RRAM. in IEEE International Reliability Physics Symposium (2019).
24. Yeo I Chu M Gi S Hwang H Lee B Stuck-at-fault tolerant schemes for memristor crossbar array-based neural networks IEEE Trans. Electron Devices 2019 66 7 10.1109/TED.2019.2914460

상세보기
25. Nguyen T Pham K Min K Hybrid circuit of memristor and complementary metal-oxide-semiconductor for defect-tolerant spatial pooling with boost-factor adjustment Materials. 2019 12 2122 10.3390/ma12132122

상세보기
26. Cui Y Ahmad S Hawkins J The HTM spatial pooler-a neocortical algorithm for online sparse distributed coding Front. Comput. Neurosci. 2017 11 111 10.3389/fncom.2017.00111 29238299

상세보기
27. Krestinskaya, O. et al., Analog HTM: memristive spatial pooler learning with backpropagation. in IEEE International Conference on Artificial Intelligence Circuits and Systems (2019).
28. Tunali M A survey of fault-tolerance algorithms for reconfigurable nano-crossbar arrays ACM Comput. Surv. 2017 50 79 1 35
29. Degraeve, R., et al., Dynamic ‘hour glass’ model for SET and RESET in HfO 2 RRAM. Symposium on VLSI Technology, T8.5 (2012).
30. Gilmer, D.C. et al., Effects of RRAM stack configuration on forming voltage and current overshoot. in IEEE International Memory Workshop (2011).
31. Sivan M All WSe 2 1T1R resistive RAM cell for future monolithic 3D embedded memory integration Nat. Commun. 2019 10 5201 10.1038/s41467-019-13176-4 31729375

상세보기
32. Nail, C. et al., Understanding RRAM endurance, retention and window margin trade-off using experimental results and simulations. in IEEE International Electron Devices Meeting (2016).
33. Kim S Lim M Kim Y Kim H Choi S Impact of synaptic device variations on pattern recognition accuracy in a hardware neural network Sci. Rep. 2018 8 2638 10.1038/s41598-018-21057-x 29422641

상세보기
34. Kim S Kim H Choi S Impact of synaptic device variations on classification accuracy in a binarized neural network Sci. Rep. 2019 9 15237 10.1038/s41598-019-51814-5 31645636

상세보기
35. Koo Y Accelerated retention test method by controlling ion migration barrier of resistive random access memory IEEE Electron Device Lett. 2015 36 3 238 240 10.1109/LED.2015.2394302

상세보기
36. Sung C Effect of a self-limited reset operation on the reset breakdown characteristics of a monolithically integrated 1T1R RRAM ECS J. Solid State Sci. Technol. 2017 6 7 440 10.1149/2.0281707jss

상세보기
37. Yu M Novel vertical 3D structure of TaO x -based RRAM with self-localized switching region by sidewall electrode oxidation Sci. Rep. 2016 6 21020 10.1038/srep21020 26884054

상세보기
38. Grossi A Impact of the precursor chemistry and process conditions on the cell-to-cell variability in 1T？1R based HfO 2 RRAM devices Sci. Rep. 2018 8 11160 10.1038/s41598-018-29548-7 30042433

상세보기
39. Calderoni, A. Sills, S., & Ramaswamy, N. Performance comparison of O-based and Cu-based ReRAM for high-density applications. in International Memory Workshop (2014).
40. Aratani, K. et al., A novel resistance memory with high scalability and nanosecond switching. in IEEE International Electron Devices Meeting (2008).
41. Woo J Introduction of WO 3 layer in a Cu-based Al 2 O 3 conductive bridge RAM system for robust cycling and large memory window IEEE J. Electron Devices Soc. 2016 4 3 10.1109/JEDS.2016.2526632

상세보기
42. Celano U Three-dimensional observation of the conductive filament in nanoscaled resistive memory devices Nano Lett. 2014 14 5 2401 2406 10.1021/nl500049g 24720425

상세보기
43. Belmonte A Origin of the current discretization in deep reset states of an Al 2 O 3 /Cu-based conductive-bridging memory, and impact on state level and variability Appl. Phys. Lett. 2014 104 23 3508 10.1063/1.4883856

상세보기
44. Yu, S. et al., Scaling-up resistive synaptic arrays for neuro-inspired architecture: challenges and prospect. in IEEE International Electron Devices Meeting (2015).
45. Chen C Endurance degradation mechanisms in TiN\Ta 2 O 5 \Ta resistive random-access memory cells Appl. Phys. Lett. 2015 106 5 053501 10.1063/1.4907573

상세보기
46. Tunali, O. et al., Logic synthesis and defect tolerance for memristive crossbar arrays. Design, Automation & Test in Europe Conference & Exhibition (DATE) (2018).
47. Paul, S., Charkraborty, R., & Bhunia, S. Defect-aware configurable computing in nanoscale crossbar for improved yield. in IEEE International On-Line Testing Symposium . (2007).
48. Truong S Pham K Min K Spatial-pooling memristor crossbar converting sensory information to sparse distributed representation of cortical neurons IEEE Trans. Nanotechnol. 2018 17 482 491 10.1109/TNANO.2018.2815624

상세보기
49. Pham K Asymmetrical training scheme of binary-memristor-crossbar-based neural networks for energy-efficient edge-computing nanoscale systems Micromachines 2019 10 141 10.3390/mi10020141

상세보기

LOADING...

활용도 분석정보

상세보기

다운로드

내보내기

활용도 Top5 논문

해당 논문의 주제분야에서 활용도가 높은 상위 5개 콘텐츠를 보여줍니다.
더보기 버튼을 클릭하시면 더 많은 관련자료를 살펴볼 수 있습니다.

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증