CrossNets: High‐Performance Neuromorphic Architectures for CMOL Circuits
10.1196/annals.1292.010
Crossref journal-article
Wiley
Annals of the New York Academy of Sciences (311)
Abstract

Abstract:The exponential,Moore's Law, progress of electronics may be continued beyond the 10‐nm frontier if the currently dominant CMOS technology is replaced by hybridCMOLcircuits combining a silicon MOSFET stack and a few layers of parallel nanowires connected by self‐assembled molecular electronic devices. Such hybrids promise unparalleled performance for advanced information processing, but require special architectures to compensate for specific features of the molecular devices, including low voltage gain and possible high fraction of faulty components. Neuromorphic networks with their defect tolerance seem the most natural way to address these problems. Such circuits may be trained to perform advanced information processing including (at least) effective pattern recognition and classification. We are developing a family of distributed crossbar network (CrossNet) architectures that permit the combination of high connectivity neuromorphic circuits with high component density. Preliminary estimates show that this approach may eventually allow us to place a cortex‐scale circuit with about 1010neurons and about 1014synapses on an approximately 10 × 10 cm2silicon wafer. Such systems may provide an average cell‐to‐cell latency of about 20 nsec and, thus, perform information processing and system training (possibly including self‐evolution after initial training) at a speed that is approximately six orders of magnitude higher than in its biological prototype and at acceptable power dissipation.

Bibliography

LIKHAREV, K., MAYR, A., MUCKRA, I., & TÜREL, Ö. (2003). CrossNets: High‐Performance Neuromorphic Architectures for CMOL Circuits. Annals of the New York Academy of Sciences, 1006(1), 146–163. Portico.

Authors 4
  1. KONSTANTIN LIKHAREV (first)
  2. ANDREAS MAYR (additional)
  3. IBRAHIM MUCKRA (additional)
  4. ÖZGÜR TÜREL (additional)
References 70 Referenced 89
  1. The International Technology Roadmap for Semiconductors 2001 Edition. Available on the Web at <http://www.public.itrs.net/Files/2001ITRS/Home.htm>.
  2. 10.1109/5.915374
  3. Likharev K.K. 2003. Electronics below 10 nm.InNano and Giga Challenges in Microelectronics. J. Greer A. Korkin & J. Labanowski Eds.: 27–68. Elsevier Amsterdam. (10.1016/B978-044451494-3/50002-0)
  4. Averin D.V. & K.K. Likharev. 1991. Single‐electronics.InMesoscopic Phenomena in Solids. B. Altshuler et al. Eds.: 173–271. Elsevier Amsterdam. (10.1016/B978-0-444-88454-1.50012-7)
  5. Grabert H. & M. Devoret Eds. 1992. Single‐Charge Tunneling. Plenum New York. (10.1007/978-1-4757-2166-9)
  6. 10.1109/5.752518
  7. 10.1109/MSPEC.2001.925276
  8. 10.1126/science.293.5531.785
  9. 10.1088/0957-4484/12/2/303
  10. 10.1111/j.1749-6632.1998.tb09884.x
  11. 10.1021/cm010165m
  12. 10.1021/ar0000612
  13. 10.1111/j.1749-6632.2002.tb03026.x
  14. Chen J. et al.2003. Molecular electronic devices. In Advanced Semiconductor and Organic Nano‐Techniques (Part III). H. Morko Ed.: 43–187. Elsevier Amsterdam. (10.1016/B978-012507060-7/50023-4)
  15. 10.1088/0957-4484/13/2/311
  16. {'issue': '226801', 'key': 'e_1_2_8_17_2', 'first-page': '1', 'article-title': 'Conductance of small molecular junctions', 'volume': '88', 'author': 'Zhitenev N.B.', 'year': '2002', 'journal-title': 'Phys. Rev. Lett.'} / Phys. Rev. Lett. / Conductance of small molecular junctions by Zhitenev N.B. (2002)
  17. 10.1038/nature00791
  18. 10.1038/nature00790
  19. 10.1126/science.285.5426.391
  20. 10.1126/science.289.5482.1172
  21. 10.1126/science.1060294 / Science / Conductance switching in single‐molecules through conformational changes by Donhauser Z.I.
  22. 10.1063/1.1521788
  23. 10.1126/science.280.5370.1716
  24. 10.1109/5.573742
  25. Mountcastle V.B. 1998. The Cerebral Cortex. Harvard University Press Cambridge.
  26. Braitenberg V. & A. Schüz. 1998. Cortex: Statistics and Geometry of Neuronal Connectivity 2nd edit. Springer Berlin. (10.1007/978-3-662-03733-1)
  27. 10.1063/1.118329
  28. 10.1002/1097-007X(200011/12)28:6<523::AID-CTA125>3.0.CO;2-R
  29. 10.1088/0957-4484/12/1/311
  30. 10.1109/16.536815
  31. 10.1063/1.121553
  32. Fölling S. Ö. Türel & K. Likharev. 2001. Single‐electron latching switches as nanoscale synapses.InProceedings of the International Joint Conference on Neural Networks: 216–221. International Neural Network Society Mount Royal New York. (10.1109/IJCNN.2001.939020)
  33. 10.1002/cta.223
  34. Türel Ö. & K.K. Likharev. 2003. CrossNets: neuromorphic networks for nanoelectronic implementation.InArtificial Neural Networks and Neural Information Processing 743–760. Springer Berlin. (10.1007/3-540-44989-2_90)
  35. Türel Ö. I. Muckra & K.K. Likharev. 2003. Possible nanoelectronic implementation of neuromorphic networks.InProceedings of the International Joint Conference on Neural Networks. 365–370. International Neural Network Society Mount Royal New York. (10.1109/IJCNN.2003.1223373)
  36. 10.1103/PhysRevLett.72.3226
  37. 10.1116/1.587625
  38. 10.1007/BF00683484
  39. 10.1103/PhysRevB.44.6199
  40. Kouwenhoven L.P. et al.1997. Electron transport in quantum dots. In Mesoscopic Electron Transfer. L. Sohn et al. Eds.: 105–215. Kluwer Dordrecht. (10.1007/978-94-015-8839-3_4)
  41. 10.1021/ja010330z
  42. 10.1021/ar9600446
  43. 10.1002/(SICI)1521-396X(199804)166:2<835::AID-PSSA835>3.0.CO;2-9
  44. 10.1002/anie.199619361
  45. 10.1021/ja017150
  46. 10.1021/cm00017a013
  47. 10.1021/jp000706f
  48. 10.1016/S0022-328X(00)94049-X
  49. 10.1021/cr9500287
  50. 10.1002/1521-3765(20011203)7:23<5118::AID-CHEM5118>3.0.CO;2-1
  51. 10.1021/ja960317s
  52. 10.1021/ja00177a027
  53. 10.1021/ja00744a016
  54. 10.1002/anie.199316431
  55. Hertz J. A. Krogh & R.G. Palmer. 1991. Introduction to the Theory of Neural Computation. Perseus Cambridge. (10.1063/1.2810360)
  56. 10.1103/PhysRevLett.57.913
  57. Fausett L. 1994. Fundamentals of Neural Networks. Prentice Hall Upper Saddle River.
  58. Haykin S. 1999. Neural Networks. Prentice Hall Upper Saddle River.
  59. Dayan P. & L.F. Abbott. 2001. Theoretical Neuroscience. MIT Press Cambridge.
  60. Rosen‐Zvi M. & I. Kanter. 2001. Training a perceptron with a discrete weight space. Phys. Rev. E64: 046109 1–6. (10.1103/PhysRevE.64.046109)
  61. 10.1016/0893-6080(88)90021-4
  62. 10.1109/31.7600
  63. 10.1109/31.7601
  64. 10.1103/PhysRevLett.61.259
  65. 10.1016/S0375-9601(00)00726-X
  66. Türel Ö. & K.K. Likharev. 2003. CMOL CrossNets: possible neuromorphic nanoelectronic circuits. Neural Information Processing Systems Meeting Vancouver Canada December 2003. Submitted for presentation. Preprint: <http://www.rsfq1.physics.sunysb.edu/~likharev/nano/Vancouver.pdf>.
  67. Fogel D.B. 1995. Evolutionary Computation. IEEE Press New York.
  68. Bäck T. 1996. Evolutionary Algorithms in Theory and Practice. Oxford Press New York. (10.1093/oso/9780195099713.003.0007)
  69. Stoica A. D. Keymeulen & J. Lohn Eds. 1999. Proceedings of the 1st NASA/DoD Workshop on Evolvable Hardware. IEEE Comp. Soc. Los Alamitos.
  70. 10.1109/5.784219
Dates
Type When
Created 21 years, 6 months ago (Feb. 19, 2004, 11:37 p.m.)
Deposited 8 months, 2 weeks ago (Dec. 15, 2024, 6:30 a.m.)
Indexed 1 day, 14 hours ago (Sept. 2, 2025, 6:36 a.m.)
Issued 21 years, 9 months ago (Dec. 1, 2003)
Published 21 years, 9 months ago (Dec. 1, 2003)
Published Online 19 years, 7 months ago (Jan. 24, 2006)
Published Print 21 years, 9 months ago (Dec. 1, 2003)
Funders 0

None

@article{LIKHAREV_2003, title={CrossNets: High‐Performance Neuromorphic Architectures for CMOL Circuits}, volume={1006}, ISSN={1749-6632}, url={http://dx.doi.org/10.1196/annals.1292.010}, DOI={10.1196/annals.1292.010}, number={1}, journal={Annals of the New York Academy of Sciences}, publisher={Wiley}, author={LIKHAREV, KONSTANTIN and MAYR, ANDREAS and MUCKRA, IBRAHIM and TÜREL, ÖZGÜR}, year={2003}, month=dec, pages={146–163} }