# Analog VLSI Implementation of Artificial Neural Networks with Supervised On-Chip Learning

- 774 Downloads
- 31 Citations

## Abstract

Analog VLSI on-chip learning Neural Networks represent a mature technology for a large number of applications involving industrial as well as consumer appliances. This is particularly the case when low power consumption, small size and/or very high speed are required. This approach exploits the computational features of Neural Networks, the implementation efficiency of analog VLSI circuits and the adaptation capabilities of the on-chip learning feedback schema.

Many experimental chips and microelectronic implementations have been reported in the literature based on the research carried out over the last few years by several research groups. The author presents and discusses the motivations, the system and circuit issues, the design methodology as well as the limitations of this kind of approach. Attention is focused on supervised learning algorithms because of their reliability and popularity within the neural network research community. In particular, the Back Propagation and Weight Perturbation learning algorithms are introduced and reviewed with respect to their analog VLSI implementation.

Finally, the author also reviews and compares the main results reported in the literature, highlighting the efficiency and the reliability of the on-chip implementation of these algorithms.

## Preview

Unable to display preview. Download preview PDF.

## References

- 1.Haykin, S.,
*Neural Networks: A Comprehensive Foundation,*2nd ed., Prentice Hall, 1999.Google Scholar - 2.Kohonen, T.,
*Self-Organization and Associative Memory.*2nd ed., Springer Verlag, 1988.Google Scholar - 3.Sheu, B. J. and Choi, J.,
*Neural Information Processing and VLSI.*Kluwer Academic Publishers, 1995.Google Scholar - 4.Cauwenberghs, G. and Bayoumi, M. (eds.),
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, 1999.Google Scholar - 5.Lippmann, R. P., “An introduction to computing with neural nets.”
*IEEE ASSP Magazine*4(2), pp. 4-22, 1987.Google Scholar - 6.Rosenblatt, F.,
*Principles of Neurodynamics: Perceptron and Theory of Brain Machines.*Washington D.C.: Spartan Books, 1962.Google Scholar - 7.Rumelhart, D. E. and McClelland, J. L.,
*Parallel Distributed Processing.*Cambridge: MIT Press, 1986.Google Scholar - 8.Murray, A. F., “Silicon implementations of neural networks,” in
*IEE Proceedings-F*138(1), February 1991.Google Scholar - 9.Anguita, D. and Valle, M., “Perspectives on dedicated hardware implementations,” in
*Proc. of the European Symposium on Artificial Neural Networks,*Bruges (Belgium), pp. 45-55, April 25-27, 2001.Google Scholar - 10.Omondi, A. R., “Neurocomputers: A dead end?”
*International Journal of Neural Systems*10(6), pp. 475-481, 2000.Google Scholar - 11.Hopfield, J. J. and Tank, D. W., “Computing with neural circuits: A model.”
*Science*233, pp. 625-633, 8 August 1986.Google Scholar - 12.Tsividis, Y. and Satyanarayana, S., “Analog circuits for variable synapse electronic neural networks.”
*Electronic Letters*2(24), pp. 1313-1314, 1987.Google Scholar - 13.Mead, C.,
*Analog VLSI and Neural Systems.*Reading MA: Addison Wesley, 1989.Google Scholar - 14.Vittoz, E. A., “Future of the analog in the VLSI environment,” in
*Proc. of ISCAS 1990,*pp. 1372-1375, 1990.Google Scholar - 15.Draghici, S., “Neural networks in analog hardware-design and implementation issues.”
*Int. Journal of Neural Systems*10(3), 2000.Google Scholar - 16.Mead, C., “Neuromorphic electronic systems,” in
*Proceedings of the IEEE*78(10), October 1990.Google Scholar - 17.Sarpeshkar, R., “Analog versus digital: Extrapolating from electronics to neurobiology.”
*Neural Computation*10, pp. 1601-1638, 1998.Google Scholar - 18.Indiveri, G., “A neuromorphic VLSI device for implementing 2-D selective attention systems.”
*IEEE Trans. on Neural Networks*12(6), pp. 1455-1463, November 2001.Google Scholar - 19.Vittoz, E., “Present and future industrial applications of bio-inspired VLSI systems.”
*Analog Integrated Systems and Signal Processing*30, pp. 173-184, 2002.Google Scholar - 20.Mahowald, M. and Douglas, R., “A silicon neuron.”
*Nature*354, pp. 19-26, December 1991.Google Scholar - 21.Andreou, A. G., “Electronic arts imitate life.”
*Nature*354, 19/26, December 1991.Google Scholar - 22.Meador, J. L. and Cole, C. S., “A low-power CMOS circuit which emulates temporal electrical properties of neurons.” In: Touretzky, S. (ed.),
*Advances in Neural Information Processing Systems.*Morgan Kaufmann Publishers, 1, 1989.Google Scholar - 23.Le Masson, S. et al., “Analog circuits for modeling biological neural networks: Design and applications.”
*IEEE Trans. on Biomedical Engineering*46(6), pp. 638-645, June 1999.Google Scholar - 24.Rasche, C. and Douglas, R. J., “Forward and backpropagation in a silicon dendrite.”
*IEEE Trans. on Neural Networks*12(2), pp. 386-393, March 2001.Google Scholar - 25.Bibyk, S. and Ismail, M., “Issues in analog VLSI and MOS techniques for neural computing.” In: Mead, C. and Ismail, M. (eds.),
*Analog VLSI Implementation of Neural Systems.*Kluwer Academic Publishers, 1989.Google Scholar - 26.Tsividis, Y., “Analog MOS integrated circuits-certain new ideas, trends, and obstacles.”
*IEEE Journal of Solid-State Circuits*SC-22(3), pp. 317-321, June 1987.Google Scholar - 27.Harris, J. G. et al., “Analog hardware implementation of continuous-time adaptive filter structures.” In: Cauwenberghs, G. and Bayoumi, M. B. (eds.),
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, 1999.Google Scholar - 28.Vittoz, E. A., “Analog VLSI signal processing: Why, where, and how?”
*Journal of VLSI Signal Processing*8, pp. 27-44, 1994.Google Scholar - 29.Cairns, G. and Tarassenko, L., “Precision issues for learning with analog VLSI multilayer perceptrons.”
*IEEEMICRO*15(3), pp. 54-56, June 1995.Google Scholar - 30.Ismail, M. and Fiez, T.,
*Analog VLSI Signal and Information Processing.*McGraw-Hill International Editors, 1994.Google Scholar - 31.Masa, P., Hoen, K. and Wallinga, H., “A high-speed analog neural processor.”
*IEEE Micro*14(3), pp. 40-50, June 1994.Google Scholar - 32.Holler, M., Tam, S., Castro, H. and Benson, R., “An electrically trainable artificial neural network (ETANN) with 10240 floating gate synapses,” in
*Proc. of IJCNN,*Washington DC, 2, pp. 191-196, 1989.Google Scholar - 33.Choi, J. and Sheu, B., “VLSI design of compact and high precision analog neural network processor,” in
*Proc. of IJCNN*2, pp. 637-641, 1992.Google Scholar - 34.Masa, P. et al., “10 mW CMOS retina and classifier for handheld, 1000 images per second optical character recognition systems,” in
*Proc. ISSCC'99,*San Francisco, 1999.Google Scholar - 35.Shima, T., Kimura, T., Kamatani, Y., Itakura, T., Fujita, Y. and Iida, T., “Neuro chips with on-chip back-propagation and/or hebbian learning.”
*IEEE Journal of Solid State Circuits*27, pp. 1868-1875, 1992.Google Scholar - 36.Valle, M., Caviglia, D. D. and Bisio, G. M., “An analog VLSI neural network with on-chip back propagation learning.”
*Analog Integrated Circuits and Signal Processing,*Kluwer AcademicGoogle Scholar - 37.Boahen, K. A. and Andreou, A. G., “A contrast sensitive silicon retina with reciprocal synapses.” In: Moody, J. E., Hanson, S. J. and Lippmann, R. P. (eds.):
*Advances in Neural Processing Systems*San Mateo, CA: Morgan Kaufmann Publishers, 4, 1992.Google Scholar - 38.Boahen, K. A., “Aretinomorphic vision system.”
*IEEE MICRO 1996,*16(5), pp. 30-39, October 1996.Google Scholar - 39.Boahen, K. A., “A retinomorphic chip with parallel pathways: Encoding increasing, on, decreasing, and off, visual signals.”
*Analog Integrated Circuits and Signal Processing*(30), pp. 121-135, 2002.Google Scholar - 40.Liu, W., Andreou, A. G. and Goldstein, M. H., “Voiced-speech representation by an analog silicon model of the auditory periphery.”
*IEEE Trans. On Neural Networks*3(3), May 1992.Google Scholar - 41.Liu, S., “Silicon photoreceptors with controllable adaptive filtering properties.” In: Cauwenberghs, G. and Bayoumi, M. B. (eds.):
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, 1999.Google Scholar - 42.Ridella, S. et al., “K-Winner machines for pattern classification.”
*IEEE Trans. on Neural Networks*12(2), pp. 371-385, March 2001.Google Scholar - 43.Cauwenberghs, G., “Learning on silicon: A survey.” In: Cauwenberghs, G. and Bayoumi, M. B. (eds.):
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, 1999.Google Scholar - 44.Hebb, D. O.,
*The Organization of Behavior, A Neuropsychological Theory.*New York: John Wiley, 1949.Google Scholar - 45.Carpenter, G. A., “Neural network models for pattern recognition and associative memory.”
*Neural Networks*2(4), pp. 243-257, 1989.Google Scholar - 46.Tarassenko, L.,
*A Guide to Neural Computing Applications.*London: Arnold Publishers, 1999.Google Scholar - 47.Hertz, J., Krogh, A. and Palmer, R. G.,
*Introduction to the Theory of the Neural Computation.*Addison-Wesley Publishing Company, 1991.Google Scholar - 48.Cichocki, A. and Unbehauen, R.,
*Neural Networks for Optimisation and Signal Processing.*John Wiley & Sons, 1993.Google Scholar - 49.Jabri, M. and Flower, B., “Weight perturbation: An optimal architecture and learning technique for analog VLSI feedforward and recurrent multilayer networks.”
*IEEE Trans. Neural Networks*3(1), pp. 154-157, 1992.Google Scholar - 50.Alspector, J. and Lippe, D., “A study of parallel perturbative gradient descent.”
*Advances in Neural Information Processing Systems (NIPS96),*pp. 803-810, 1996.Google Scholar - 51.Jabri, M. A., Coggins, R. F. and Flower, B. G.,
*Adaptive Analog VLSI Neural Systems.*Chapman & Hall, 1996.Google Scholar - 52.Cauwenberghs, G., “A fast stochastic error-descent algorithm for supervised learning and optimization.”
*Advances in Neural Information Processing Systems 5 (NIPS5),*pp. 244-251, 1993.Google Scholar - 53.Alspector, J., Meir, R., Yuhas, B., Jayakumar, A. and Lippe, D., “Aparallel gradient descent method for learning in analog VLSI neural networks.”
*Advances in Neural Information Processing Systems 5 (NIPS5),*pp. 836-844, 1993.Google Scholar - 54.Reyneri, L. M. and Filippi, E., “An analysis on the performance of silicon implementations of backpropagation algorithms for artificial neural networks.”
*IEEE Trans. on Computers*12, pp. 1380-1389, 1991.Google Scholar - 55.Caviglia, D. D., Valle, M., Rossi, A., Vincentelli, M., Bo, G. M., Colangelo, P., Pedrazzi, P. and Colla, A., “Feature extraction circuit for optical character recognition.”
*Electronics Letters*30(10), pp. 769-760, 1994.Google Scholar - 56.Jacobs, B. A., “Increased rates of convergence through learning rate adaptation.”
*Neural Networks*4, pp. 295-307, 1988.Google Scholar - 57.Tollenaere, T., “SuperSAB: Fast adaptive back propagation with good scaling properties.”
*Neural Networks*3, pp. 561-573, 1990.Google Scholar - 58.Edwards, P. J. and Murray, A. F., “Analogue imprecision in MLP training.”World Scientific Publishing Co. Pte. Ltd., 1996.Google Scholar
- 59.Valle, M., Caviglia, D. D., Donzellini, G., Mussi, A., Oddone, F. and Bisio, G. M., “A neural computer based on an analog VLSI neural network,” in
*Proc. of the International Conference on Artificial Neural Network, ICANN94*2, pp. 1339-1342, 1994.Google Scholar - 60.Bo, G. M., Caviglia, D. D., Chiblé, H. and Valle, M., “A circuit architecture for analog on-chip back propagation learning with local learning rate adaptation.”
*Analog Integrated Circuits and Signal Processing*18(2/3), pp. 163-174, 1999.Google Scholar - 61.Bo, G. M., Caviglia, D. D., Chiblé, H. and Valle, M., “A self-learning analog neural processor.” To be published on IEICE Trans. on Fundamentals, 2002.Google Scholar
- 62.Murray, A. F. and Tarassenko, L.,
*Analogue Neural VLSI: A Pulse Stream Approach.*Chapman & Hall, 1994.Google Scholar - 63.Kondo, Y. and Sawada, Y., “Functional abilities of a stochastic logic neural network.”
*IEEE Trans. on Neural Networks*3(3), pp. 434-443, May 1992.Google Scholar - 64.Andreou, A. G. et al., “Current-mode subthreshold MOS circuits for analog VLSI neural systems.”
*IEEE Trans. on Neural Networks*2(2), March 1991.Google Scholar - 65.Andreou, A. G. and Boahen, K. A., “Translinear circuits in subthreshold CMOS.”
*Analog Integrated Circuits and Signal Processing*9, pp. 141-166, 1996.Google Scholar - 66.Alspector, J. et al., “VLSI architectures for neural networks.” In: Antognetti, P. and Milutinovic V. (eds.):
*Neural Networks: Concepts, Applications and Implementations*I, Prentice Hall Publisher, pp. 180-215, 1991.Google Scholar - 67.Annema, A.,
*Feed-Forward Neural Networks: Vector Decomposition Analysis, Modelling and Analog Implementation.*Kluwer Academic Publishers, 1995.Google Scholar - 68.Sarné, G. M. L. and Pastorino, M. N., “Application of neural network for the simulation of the traffic flows in a real transportation network,” in
*Proc. of the International Conference on Artificial Neural Networks ICANN94,*pp. 831-833, 1994.Google Scholar - 69.Bo, G. M., Caviglia, D. D. and Valle, M., “An analog VLSI neural architecture for handwritten numeric character recognition,” in
*Proc. of the International Conference on Artificial Neural Networks ICANN95, Industrial Conference,*Paris, France, 1995.Google Scholar - 70.Bourlard, H. and Morgan, N., “Hybrid connectionist models for continuous speech recognition.” In: Lee, C., Soong, F. K. and Paliwal, K. K. (eds.):
*Automatic Speech and Speaker Recognition.*Kluwer Academic Publishers, pp. 259-283, 1996.Google Scholar - 71.Diotalevi, F., Bo, G. M., Caviglia, D. D. and Valle, M., “Evaluation and validation of local and adaptive weight perturbation learning algorithms for optical character recognition applications,” in
*Proc. of the Third International ICSC Symposia on Intelligent Industrial Automation (IIA'99) and Soft Computing (SOCO'99),*Genova, pp. 508-512, 1-4 June, 1999.Google Scholar - 72.Diotalevi, F., Valle, M. and Caviglia, D. D., “An adaptive local rate technique for hierarchical feed forward architectures.”
*IEEE-INNS-ENNS International Joint Conference on Neural Networks.*Piscataway, NJ, Como, Italy: IEEE Press, 24-27 July 2000.Google Scholar - 73.Diotalevi, F., “Analog microelectronic supervised learning systems.” Ph.D. Thesys, http://www.micro.dibe.unige.it/works/FDiotalevi_PHD_Thesis.zip.Google Scholar
- 74.Diotalevi, F. and Valle, M., “Weight perturbation learning algorithm with local learning rate adaptation for the classification of remote-sensing images.”
*ESANN'01,*Bruges (Belgium): D-Facto Publisher, pp. 217-222, 25-27 April 2001. (ISBN 2-930307-01-3).Google Scholar - 75.Cauwenberghs, G., “An analog VLSI recurrent neural network learning a continuous-time trajectory.”
*IEEE Transaction on Neural Networks*7(2), pp. 346-361, 1996.Google Scholar - 76.Hollis, P. W. et al., “The effects of precision constraints in a backpropagation learning network.”
*Neural Computation*2, pp. 363-373, 1990.Google Scholar - 77.Anguita, D. et al., “Worst case analysis of weight inaccuracy effects in multilayer perceptrons.”
*IEEE Trans. on Neural Networks*10(2), pp. 415-418, March 1999.Google Scholar - 78.Montalvo, A. J. et al., “An analog VLSI neural network with on-chip perturbation learning.”
*IEEE Journal of Solid State Circuits,*32(4), pp. 535-543, April 1997.Google Scholar - 79.Montalvo, A. J. et al., “Towards a general-pourpose analog VLSI neural network with on-chip learning.”
*IEEE Trans. on Neural Networks*8(2), pp. 413-423, March 1997.Google Scholar - 80.Horio, Y. et al., “Analog memories for VLSI neurocomputing.”
*IEEE International Symposium on Circuits and Systems*4, pp. 2986-2989, 1990.Google Scholar - 81.Hollis, P. W. and Paulos, J. J., “Artificial neural networks using MOS analog multiplier.”
*IEEE JSSCs*25(3), pp. 849-855, June 1990.Google Scholar - 82.van Der Spiegel, J. V. et. al., “An analog neural computer with modular architecture for real-time dynamic computations.”
*IEEE JSSCs*22(1), pp. 82-91, January 1992.Google Scholar - 83.Kramer, A. et al., “Flash-based programmable nonlinear capacitor for switched-capacitor implementations of neural networks.”
*IEDM Tech. Dig.,*pp. 17.6.1-17.6.4, December 1994.Google Scholar - 84.Pavan, P., Bez, R., Olivo, P. and Zanoni, E., “Flash memory cells-an overview,” in
*Proceedings of the IEEE*85(8), pp. 1248-1271, 1997.Google Scholar - 85.Harrison, R. R. et al., “Special issue on floating-gate devices, circuits and systems.”
*IEEE Trans. on Circuits and Systems II: Analog and Digital Signal Processing*48(1), January 2001.Google Scholar - 86.Kim, K., Lee, K., Jung, T. and Suh, K., “An 8-bit-resolution, 360-
*u*s write time non-volatile analog memory based on differentially balanced constant-tunnelling-current scheme (DBCS).”*IEEE Journal of Solid State Circuits*33(11), pp. 1758-1762, 1998.Google Scholar - 87.Harrison, R. R. et al., “A CMOS programmable analog memory-cell array using floating-gate circuits.”
*IEEE Trans. on Circuits and Systems II: Analog and Digital Signal Processing*48(1), pp. 4-11, January 2001.Google Scholar - 88.Murray, A. F., “Pulse arithmetic in VLSI neural networks.”
*IEEE Micro*9(6), pp. 64-74, 1989.Google Scholar - 89.Schwartz, D. B. et al., “A programmable analog neural network chip.”
*IEEE Journal of Solid State Circuits*24(2), pp. 313-319, 1989.Google Scholar - 90.Kub, F. J. et al., “Programmable analog vector-matrix multipliers.”
*IEEE Journal of Solid State Circuits*25(1), pp. 207-214, February 1990.Google Scholar - 91.Tsividis, Y.,
*Mixed Analog-Digital VLSI Devices and Technology: An Introduction.*McGraw-Hill, 1995.Google Scholar - 92.Sheu, B. J., Shieh, J. and Patil, M., “Modelling charge injection in MOS analog switch.”
*IEEE Transaction on Circuits and Systems*34(2), pp. 214-216, 1987.Google Scholar - 93.Vittoz, E. et al., “Analog storage of adjustable synaptic weights.” In: Ramacher, U. (ed.)
*Introduction to VLSI Design of Neural Networks.*Kluwer Academic Publisher, 1991.Google Scholar - 94.Castello, R., Caviglia, D. D., Franciotta, M. and Montecchi, F., “Self refreshing analogue memory cell for variable synaptic weights.”
*IEE Electronics Letters*27(20), pp. 1871-1872, 1991.Google Scholar - 95.Hochet, B. et al., “Implementation of a learning Kohonen neuron based on a new multilevel storage technique.”
*IEEE Journal of Solid State Circuits*26(3), pp. 262-267, March 1991.Google Scholar - 96.Cauwenberghs, G. and Yariv, A., “Fault-tolerant dynamic multilevel storage in analog VLSI.”
*IEEE Trans. on Circuits and Systems II*41(2), pp. 827-829, 1994.Google Scholar - 97.Ehlert, M. and Klair, H., “A 12-bit medium-time analog storage device in a CMOS standard process.”
*IEEE Journal of Solid State Circuits*33(7), pp. 1139-1143, 1998.Google Scholar - 98.Pelgrom, M. J. M., Aad, L., Duinmaijer, C. J. and Welbers, A. P. G., “Matching properties of MOS transistors.”
*IEEE Journal of Solid-State Circuits*24(5), pp. 1433-1440, October 1989.Google Scholar - 99.Kinget, P. and Steyaert, M., “Analog VLSI integration of massive parallel processing systems.” Kluwer Academic Publishers, November 1996.Google Scholar
- 100.Shoval, A., Johns, D. A. and Snelgrove, W. M., “Comparison of DC offset effects in four LMS adaptive algorithms.”
*IEEE Transaction on Circuits and Systems II*42(3), pp. 176-185, 1995.Google Scholar - 101.Satyanarayana, S., Tsividis, Y. P. and Graf, H. P., “A reconfigurable VLSI neural network.”
*IEEE Journal of Solid State Circuits*27(1), pp. 67-81, 1992.Google Scholar - 102.Dolenko, B. K. and Card, H. C., “Tolerance to analog hardware of on-chip learning in backpropagation networks.”
*IEEE Trans. on Neural Networks*6(5), pp. 1045-1052, 1995.Google Scholar - 103.Morie, T., “Analog VLSI implementation of self learning neural networks.” In: Cauwenberghs, G. and Bayoumi, M. (eds.):
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, 1999.Google Scholar - 104.Alhalabi, B. A., Bayoumi, M. B. and Maaz, B., “Mixed-mode programmable and scalable architecture for on-chip learning.”
*Analog Integrated Circuits and Signal Processing*(18), pp. 175-194, 1999.Google Scholar - 105.Murray, A. F. and Woodburn, R., “The prospects for analogue neural VLSI.”
*International Journal of Neural Systems*8, pp. 559-579, October/December 1997.Google Scholar - 106.Lehmann, T. and Woodburn, R., “Biologically-inspired learning in pulsed neural networks.” In: Cauwenberghs, G. and Bayoumi, M. (eds.):
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, pp. 105-130, 1999.Google Scholar - 107.Gray, P. R. and Meyer, R. G.,
*Analysis and Design of Analog Integrated Circuits,*2nd Edition, McGraw Hill, 1984.Google Scholar - 108.Osa, J. I., Carlosena, A. and Lopez-Martin, A. J., “MOSFET-C filter with on-chip tuning and wide programming range.”
*IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing*48(10), pp. 944-951, October 2001.Google Scholar - 109.Tsividis, Y., “Integrated continuous-time filter design-an overview.”
*IEEE Journal of Solid State Circuits*29(3), pp. 166-176, March 1994.Google Scholar - 110.Andreou, A. G. and Furth, P. M., “An information theoretic framework for comparing the bit energy of signal representations at the circuit level.” In: Sanchez-Sinencio, E. and Andreou, A. G. (eds.):
*Low-Voltage Low-Power Integrated Circuits and Systems.*IEEE Press, 1999.Google Scholar - 111.Jespers, P. G. A.,
*Integrated Converters.*Oxford University Press, 2001.Google Scholar - 112.Kramer, A. H., “Array-based computation: Principles, advantages and limitations,” in
*Proc. of Microneuro'96,*pp. 68-79, 1996.Google Scholar - 113.Frantz, G., “Digital signal processor trends.”
*IEEE Micro*20(6), pp. 52-59, November/December 2000.Google Scholar - 114.Cauwenberghs, G., Bayoumi, M. and Sanchez-Sinencio, E. (eds.),
*Special Issue on Learning on Silicon, Analog Integrated Circuits and Signal Processing.*Kluwer Academic Publisher, 18(2-3), 1999.Google Scholar - 115.Lehmann, T., “Hardware learning in analog VLSI neural networks.” Ph.D. Thesis, Electronics Institute, Technical University of Denmark, 1994. (http://eivind.imm.dtu.dk/publications/phdthesis.html).Google Scholar
- 116.Berg, Y., Sigvartsen, R. L., Lande, T. S. and Abusland, A., “An analog feed-forward neural network with on-chip learning.”
*Analog Integrated Circuits and Signal Processing*9, pp. 65-75, 1996.Google Scholar - 117.Bo, G. M., Chiblé, H., Caviglia, D. D. and Valle, M., “Analog VLSI on-chip learning neural network with learning rate adaptation.” In: Cauwenberghs, G., Bayoumi, M. A. (eds.):
*Learning on Silicon-Adaptive VLSI Neural Systems.*Kluwer Academic Publishers, pp. 305-330, June 1999.Google Scholar - 118.Lu, C., Shi, B. and Chen, L., “An on-chip BP learning neural network with ideal neuron characteristics and learning rate adaptation.”
*Analog Integrated Circuits and Signal Processing*31, pp. 55-62, 2002.Google Scholar - 119.http://dspvillage.ti.com/docs/dspproducthome.jhtmlGoogle Scholar
- 120.Kramer, A. H., “Array-based analog computation,” in
*IEEE Micro 1996*16(5), pp. 20-29, October 1996.Google Scholar - 121.Genov, R. and Cauwenberghs, G., “Charge-mode parallel architecture for vector-matrix multiplication.”
*IEEE Trans. on CAS II-Analog and Digital Signal Processing*48(10), pp. 930-936, October 2001.Google Scholar - 122.Caviglia, D. D., Valle, M. and Bisio, G. M., “Effects of weight discretization on the back propagation learning method: Algorithm design and hardware realization.” in
*Proc. of the IEEE-INNS International Joint Conference on Neural Networks IJCNN,*Ann Arbor, MI, San Diego (USA), pp. II 631-II 637, June 17-21, 1990.Google Scholar - 123.Valle, M., Caviglia, D. D. and Bisio, G. M., “Back-propagation learning algorithms for analog VLSI implementation, in VLSI for neural networks and artificial intelligence,” in
*Proc. International Workshop on VLSI for Neural Networks and Artificial Intelligence,*New York, Oxford UK: Plenum Press, 2-4 September 1992; In: Delgado-Frias, J. D. and Moore, W. R. (eds.), 1994, pp. 35-44. (ISBN: 0-306-44722-3).Google Scholar - 124.Aslam-Siddiqi, A. et al., “A 16 × 16 nonvolatile programmable analog vector-matrix multiplier.”
*IEEE Journal of Solid-State Circuits,*33(10), pp. 1502-1509, October 1998.Google Scholar - 125.Diotalevi, F., Valle, M., Bo, G. M., Biglieri, E. and Caviglia, D. D., “Analog CMOS current mode neural primitives.” In
*Proc. ISCAS'2000, IEEE International Symposium on Circuits and Systems,*Geneva, Switzerland, 28-31 May, 2000, pp. I 419-I 422. (ISBN: 0-7803-5485-0).Google Scholar - 126.Diotalevi, F., Valle, M., Bo, G. M., Biglieri, E. and Caviglia, D. D., “An analog on-chip learning circuit architecture of the weight perturbation algorithm.”
*ISCAS'2000, IEEE International Symposium on Circuits and Systems,*Geneva, Switzerland, 28-31 May, 2000, pp. II 717-II 720. (ISBN: 0-7803-5485-0).Google Scholar