Abstract
The memory capacity, computational power, communication bandwidth, energy consumption, and physical size of the brain all tend to scale with the number of synapses, which outnumber neurons by a factor of 10,000. Although progress in cortical simulations using modern digital computers has been rapid, the essential disparity between the classical von Neumann computer architecture and the computational fabric of the nervous system makes large-scale simulations expensive, power hungry, and time consuming. Over the last three decades, CMOS-based neuromorphic implementations of “electronic cortex” have emerged as an energy efficient alternative for modeling neuronal behavior. However, the key ingredient for electronic implementation of any self-learning system—programmable, plastic Hebbian synapses scalable to biological densities—has remained elusive. We demonstrate the viability of implementing such electronic synapses using nanoscale phase change devices. We introduce novel programming schemes for modulation of device conductance to closely mimic the phenomenon of Spike Timing Dependent Plasticity (STDP) observed biologically, and verify through simulations that such plastic phase change devices should support simple correlative learning in networks of spiking neurons. Our devices, when arranged in a crossbar array architecture, could enable the development of synaptronic systems that approach the density (∼1011 synapses per sq cm) and energy efficiency (consuming ∼1pJ per synaptic programming event) of the human brain.
- Abbott, L. F. and Blum, K. I. 1994. Functional significance of long-term potentiation for sequence learning and prediction. Cerebral Cortex 6, 406--416.Google ScholarCross Ref
- Abbott, L. F. and Nelson, S. B. 2000. Synaptic plasticity: taming the beast. Nat. Neurosci. 3 Suppl, 1178--1183.Google ScholarCross Ref
- Adler, D., Henisch, H. K., and Mott, S. N. 1978. The mechanism of threshold switching in amorphous alloys. Rev. Mod. Phys. 50, 2, 209--220.Google ScholarCross Ref
- Allen, C. and Stevens, C. F. 1994. An evaluation of causes for unreliability of synaptic transmission. Proc. Nat. Acad. Sci. 91, 10380--10383.Google ScholarCross Ref
- Ananthanarayanan, R., Esser, S. K., Simon, H. D., and Modha, D. S. 2009. The cat is out of the bag cortical simulations with 109 neurons, 1013 synapses. In Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis. ACM, 1--2. Google ScholarDigital Library
- Ananthanarayanan, R. and Modha, D. S. 2010. Anatomy of a cortical simulator. In Proceedings of the ACM/IEEE Conference on High Performance Networking and Computing. 1--2. Google ScholarDigital Library
- Arthur, J. V. and Boahen, K. 2006. Learning in silicon: Timing is everything. In Advances in Neural Information Processing Systems 17, B. Sholkopf and Y. Weiss, Eds., MIT Press, 281--1185.Google Scholar
- Bell, C. C., Han, V. Z., Sugawara, Y., and Grant, K. 1997. Synaptic plasticity in a cerebellum-like structure depends on temporal order. Nature 387, 278--281.Google ScholarCross Ref
- Bi, G. Q. and Poo, M. M. 1998. Synaptic modifications in cultured hippocampal neurons: dependence on spike timing, synaptic strength, and postsynaptic cell type. J. Neurosci. 18, 10464--10472.Google ScholarCross Ref
- Blum, K. I. and Abbott, L. F. 1996. A model of spatial map formation in the hippocampus of the rat. Neural Computation 8, 85--93. Google ScholarDigital Library
- Bofill, A., Murray, A. F., and Thompson, D. P. 2001. Circuits for VLSI implementation of temporally asymmetric Hebbian learning. In Advances in Neural Information Processing Systems 14. MIT Press, 1091--1098.Google Scholar
- Boniardi, M., Ielmini, D., Lavizzari, S., Lacaita, A., Redaelli, A., and Pirovano, A. 2009. Statistical and scaling behavior of structural relaxation effects in phase-change memory (pcm) devices. In Proceedings of the IEEE International.Reliability Physics Symposium. 122--127.Google Scholar
- Brader, J. M., Senn, W., and Fusi, S. 2007. Learning real-world stimuli in a neural network with spike-driven synaptic dynamics. Neural Computation 19, 2881--912. Google ScholarDigital Library
- Breitwisch, M., Nirschl, T., et al. 2007. Novel lithography-independent pore phase change memory. In Proceedings of the IEEE Symposium on VLSI Technology. 100--101.Google ScholarCross Ref
- Burr, G. W., Breitwisch, M. J., et al. 2010. Phase change memory technology. J. Vac. Sci. Technol., B 28.Google ScholarCross Ref
- Choi, H., Jung, H., Lee, J., Yoon, J., Park, J., Jun Seong, D., Lee, W., Hasan, M., Jung, G.-Y., and Hwang, H. 2009. An electrically modifiable synapse array of resistive switching memory. Nanotechnology 20, 34, 345201.Google ScholarCross Ref
- Chou, S. Y., Krauss, P. R., and Renstrom, P. J. 1996. Imprint lithography with 25-nanometer resolution. Science 272, 5258, 85--87.Google Scholar
- Dan, Y. and Poo, M. M. 2004. Spike timing-dependent plasticity of neural circuits. Neuron 44, 23--30.Google ScholarCross Ref
- Debanne, D., Gahwiler, B. H., and Thompson, S. M. 1998. Long-term synaptic plasticity between pairs of individual CA3 pyramidal cells in rat hippocampal slice cultures. J. Physiol. (Lond.) 507 (Pt 1), 237--247.Google ScholarCross Ref
- Djurfeldt, M., Lundqvist, M., Johansson, C., Rehn, M., Ekeberg, O., and Lansner, A. 2008. Brain-scale simulation of the neocortex on the IBM blue gene/l supercomputer. IBM J. Res. Dev. 52, 31--41. Google ScholarDigital Library
- Egger, V., Feldmeyer, D., and Sakmann, B. 1999. Coincidence detection and efficacy changes in synaptic connections between spiny stellate neurons of the rat barrel cortex. Nat. Neurosci. 2, 10981105.Google ScholarCross Ref
- Feldman, D. E. 2000. Timing-based LTP and LTD at vertical inputs to layer II/III pyramidal cells in rat barrel cortex. Neuron 27, 45--56.Google ScholarCross Ref
- Friesz, A. K., Parker, A. C., Zhou, C., Ryu, K., and Sanders, J. M. 2007. A biomimetic carbon nanotube synapse circuit. In Proceedings of the Biomedical Engineering Society (BMES) Annual Fall Meeting.Google Scholar
- Gallagher, W. J. and Parkin, S. S. P. 2006. Development of the magnetic tunnel junction MRAM at IBM: From first junctions to a 16-mb MRAM demonstrator chip. IBM J. Res. Dev. 50, 1, 5--23. Google ScholarDigital Library
- Gao, C. and Hammerstrom, D. 2007. Cortical models onto cmol and cmos: Architectures and performance/price. IEEE Trans. Circuits Syst. Regul. Pap. 54, 11, 2502--2515.Google ScholarCross Ref
- Gerstner, W., Kempter, R., Van Hemmen, J. L., and Wagner, H. 1996. A neuronal learning rule for sub-millisecond temporal coding. Nature 383, 76--78.Google ScholarCross Ref
- Hasegawa, T., Ohno, T., Terabe, K., Tsuruoka, T., Nakayama, T., Gimzewski, J. K., and Aono, M. 2010. Learning abilities achieved by a single solid-state atomic switch. Adv. Mater. 22, 16, 1831--1834.Google ScholarCross Ref
- Hopfield, J. J. and Brody, C. D. 2004. Learning rules and network repair in spike-timing-based computation networks. Proc. Nat. Acad. Sci. 101, 1, 337--342.Google ScholarCross Ref
- Hynna, K. and Boahen, K. 2007. Silicon neurons that burst when primed. In Proceedings of the IEEE International Symposium on Circuits and Systems. 3363--3366.Google Scholar
- Indiveri, G., Chicca, E., and Douglas, R. 2006. A VLSI array of low-power spiking neurons and bistable synapses with spike-timing dependent plasticity. IEEE Trans. Neural Networks 17, 1, 211--221. Google ScholarDigital Library
- Izhikevich, E. M. 2006. Polychronization: computation with spikes. Neural Comput 18, 245--282. Google ScholarDigital Library
- Izhikevich, E. M. and Edelman, G. M. 2008. Large-scale model of mammalian thalamocortical systems. Proc. Nat. Acad. Sci. 105, 9, 3593--3598.Google ScholarCross Ref
- Jo, S. H., Chang, T., Ebong, I., Bhadviya, B. B., Mazumder, P., and Lu, W. 2010. Nanoscale memristor device as synapse in neuromorphic systems. Nano Lett. 10, 4, 1297--1301.Google ScholarCross Ref
- Kandel, E. R. 2000. Nobel Lectures, Physiology or Medicine 1996--2000.Google Scholar
- Lai, S. and Lowrey, T. 2001. Oum: A 180nm nonvolatile memory cell element technology for stand alone and embedded applications. In Proceedings of the International Electron Devices Meeting, IEDM Technical Digest. 36.5.1--36.5.4.Google Scholar
- Lazzaro, J. 1992. Low-power silicon spiking neurons and axons. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS'92). Vol. 5. 2220--2223.Google ScholarCross Ref
- Likharev, K. and Strukov, D. 2005. Cmol - devices, circuits, and architectures. In Introducing Molecular Electronics, G. Cuniberti, K. Richter, and G. Fages, Eds., Springer, 447--477.Google Scholar
- Maass, W. and Natschlager, T. 2000. A model for fast analog computation based on unreliable synapses. Neural Comput. 12, 1679--1704. Google ScholarDigital Library
- Magee, J. C. and Johnston, D. 1997. A synaptically controlled, associative signal for Hebbian plasticity in hippocampal neurons. Science 275, 5297, 209--213.Google Scholar
- Markram, H., Lübke, J., Frotscher, M., and Sakmann, B. 1997. Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs. Science 275, 5297, 213--215.Google Scholar
- Mead, C. 1990. Neuromorphic electronic systems. Proc. IEEE 78, 10, 1629--1636.Google ScholarCross Ref
- Mehta, M. R., Quirk, M. C., and Wilson, M. A. 2000. Experience-dependent asymmetric shape of hippocampal receptive fields. Neuron 25, 707--715.Google ScholarCross Ref
- Minai, A. A. and Levy, W. B. 1993. Sequence learning in a single trial. In Proceedings of the INNS World Congress on Neural Networks. Erlbaum, 505--508.Google Scholar
- Nirschl, T., Phipp, J., et al. 2007. Write strategies for 2 and 4-bit multi-level phase-change memory. In Proceedings of the IEEE International Electron Devices Meeting. 461--464.Google ScholarCross Ref
- Ovshinsky, S. R. 1968. Reversible electrical switching phenomena in disordered structures. Phys. Rev. Lett. 21, 20, 1450--1453.Google ScholarCross Ref
- Ovshinsky, S. R. 2004a. Optical cognitive information processing—A new field. Japanese J. Appl. Phys. 43, 7B, 4695--4699.Google ScholarCross Ref
- Ovshinsky, S. R. 2004b. The ovonic cognitive computer: A new paradigm. In EPCOS Library.Google Scholar
- Pirovano, A., Lacaita, A., Pellizzer, F., Kostylev, S., Benvenuti, A., and Bez, R. 2004. Low-field amorphous state resistance and threshold voltage drift in chalcogenide materials. IEEE Trans. Electron Devices 51, 5, 714--719.Google ScholarCross Ref
- Rao, R. P. N. and Sejnowski, T. J. 2001. Spike-timing-dependent hebbian plasticity as temporal difference learning. Neural Comput. 13, 2221--2237. Google ScholarDigital Library
- Raoux, S. and Wuttig, M. 2009. Phase Change Materials. Springer.Google Scholar
- Redaelli, A., Pirovano, A., Benvenuti, A., and Lacaita, A. L. 2008. Threshold switching and phase transition numerical models for phase change memory simulations. J. Appl. Phys. 103, 11, 111101.Google ScholarCross Ref
- Roberts, P. D. 1999. Computational consequences of temporally asymmetric learning rules: I. Differential Hebbian learning. J. Comput. Neurosci. 7, 235--246.Google ScholarCross Ref
- Rochester, N., Holland, J. H., Haibt, L. H., and Duda, W. L. 1956. Tests on a cell assembly theory of the action of the brain using a large digital computer. IRE Trans. Inf. Theory PGIT-2, 3, 80--93.Google ScholarCross Ref
- Scott, J. and Bozano, L. 2007. Nonvolatile memory elements based on organic materials. Adv. Mater. 19, 11, 1452--1463.Google ScholarCross Ref
- Silver, R., Boahen, K., Grillner, S., Kopell, N., and Olsen, K. L. 2007. Neurotech for neuroscience: Unifying concepts, organizing principles, and emerging tools. J. Neurosci. 27, 44, 11807--11819.Google ScholarCross Ref
- Song, S., Miller, K. D., and Abbott, L. F. 2000. Competitive Hebbian learning through spike-timingdependent synaptic plasticity. Nat. Neurosci. 3, 919--926.Google ScholarCross Ref
- Tour, J., Van Zandt, W., Husband, C., Husband, S., Wilson, L., Franzon, P., and Nackashi, D. 2002. Nanocell logic gates for molecular computing. IEEE Trans. Nanotechnol. 1, 2, 100--109. Google ScholarDigital Library
- Waser, R., Dittmann, R., Staikov, G., and Szot, K. 2009. Redox-based resistive switching memories nanoionic mechanisms, prospects, and challenges. Adv. Mater. 21, 25, 2632--2663.Google ScholarCross Ref
- Wuttig, M. and Yamada, N. 2007. Phase-change materials for rewriteable data storage. Nat. Mater. 6, 5258, 824--832.Google Scholar
- Zhang, L. I., Tao, H. W., Holt, C. E., Harris, W. A., and Poo, M. 1998. A critical window for cooperation and competition among developing retinotectal synapses. Nature 395, 37--44.Google ScholarCross Ref
Index Terms
- Nanoscale electronic synapses using phase change devices
Recommendations
Computer Simulation of Vestibuloocular Reflex Motor Learning Using a Realistic Cerebellar Cortical Neuronal Network Model
Neural Information ProcessingAbstractThe vestibuloocular reflex (VOR) is under adaptive control to stabilize our vision during head movements. It has been suggested that the acute VOR motor learning requires long-term depression (LTD) and potentiation (LTP) at the parallel fiber – ...
2007 Special Issue: Fading memory and time series prediction in recurrent networks with different forms of plasticity
We investigate how different forms of plasticity shape the dynamics and computational properties of simple recurrent spiking neural networks. In particular, we study the effect of combining two forms of neuronal plasticity: spike timing dependent ...
Learning Precise Spike Timings with Eligibility Traces
Artificial Neural Networks and Machine Learning – ICANN 2020AbstractRecent research in the field of spiking neural networks (SNNs) has shown that recurrent variants of SNNs, namely long short-term SNNs (LSNNs), can be trained via error gradients just as effective as LSTMs. The underlying learning method (e-prop) ...
Comments