Supervised Competitive Learning (SCL) assembles a set of learning modules into a supervised learning system to address the stability-plasticity dilemma. Each learning module acts as a similarity detector for a prototype, and includes prototype resetting (akin to that of ART) to re spond to new prototypes. SCL has usually employed backpropagation networks as the learning modules. It has been tested with two feature abstractors: about 30 energy-based features, and a com bination of energy-based and graphical features (about 60). About 75 subjects have been involved. In recent testing (15 college students), SCL recognized 99 % (energy features only) of test digits, 91 % (energy) and 96.6% (energy/graphical) of test letters, and 85% of test gestures (energy/graphical). SCL has also been tested with fuzzy sets as learning modules for recognizing handwritten digits and handwritten gestures, recognizing 97% of test digits and 91% of test gestures.
Get full access to this article
View all access options for this article.
References
1.
Aleksander, I. and H. Morton.1990. An Introduction to Neural ComputingChapman and Hall
2.
Baran, N.1992. "The Outlook for Pen Computing", Byte (September) 159-164.
3.
Becker, S. and Y le Cun.1988. "Improving the Convergence of BackPropagation Learning with Second Order Methods", in Proceedings of the 1988 Connectconast Models Summer School (CMU, June 17-26, 1988) , D. Touretzky, G. Hinton and T Sejnowski, eds , San Mateo, CA: Morgan Kaufmann Publishers, pp. 29-37.
4.
Carpenter, G.A. and S. Grossberg.1987. "ART2: Self-Organizing of Stable Category Recognition Codes for Analog Input Patterns", Applied Optics , 4919-4930.
5.
Carpenter, G.A. and S. Grossberg.1988. "The ART of Adaptive Pattern Recogmtion by a Self-Organizing Neural Network", Computer, 21:3, 77-88.
6.
Carr, R. and D. Shafer.1991. The Power of PenPoint Reading, MA: Addison-Wesley Publishing Company, Inc.
7.
Fahlman, S.E.1988"Faster Learning Variations on Backpropagation An Empirical Study'; in Proceedings of the 1988 Connectionist Models Summer School (CMU, June 17-26, 1988), D Touretzsky, G. Hinton and T. Sejnowski, eds., San Mateo, CA. Morgan Kaufmann, 1988, pp. 38-51.
8.
Fukushima, K.1988. "Neocognitron: A Hierarchical Neural Network Capable of Visual Pattern Recognition", Neural Networks , 1:119-130.
9.
Fuller, T.1987. Microcomputers in Production and Inventory ManagementHomewood, IL: Dow Jones-Irwm.
10.
Fuller, T.1991. "The Kumon Approach to Learning Mathematics: An Educator's Perspective", Technical Report WUCS 91-49, WashingtonUniversity in St Louis.
11.
Fuller, T.1992. "Energy-Related Feature Abstraction for Handwritten Digit Recognition", Proceedings of the Fourth Midwest Artificial Intelligence and Cognitive Science Conference, May 1992, pp. 71-76.
12.
Gibbs, M.1993"Handwriting Recognition: A Comprehensive Comparison" , Pen (March/April)·31-35.
13.
GO Corporation.1991PenPomt API ReferenceFoster City, CA: GO Corporation.
14.
Grossberg, S.1986"Competitive Learning: From Interactive Activation to Adaptive Resonance", Cognitive Science, 11:23-63.
15.
Grothe, R.1991. Private communication with the authors, October, 1991.
16.
Herbst, N.M. and C.N. Liu.1982"Automatic Signature Verification", in Computer Analysis and Perception, Volume 1: Visual Signals , C Y Suen and R. De Mori, eds., Boca Raton, FL· CRC Press, Inc , pp. 83-105.
Kalman, B.L. and S.C. Kwasny, 1991. "A Supenor Error Function for Training Neural Nets", Proceedings of the International Joint Conference on Neural Networks . Seattle, Washington, 1991, Vol 2, pp. 49-52
19.
Kalman, B.L. and S.C. Kwasny.1992"Why Tanh: Choosing a Sigmoidal Function", Proceedings of International Joint Conference on Neural Networks , Baltimore, MD, June 1992.
20.
Kimura, T.D.1990. "An Algebraic Proof for Backpropagation in Acyclic Neural Networks'; Proceedings of International Jomt Conference on Neural Networks1990, San Diego, CA, June 1990, pp 554-557
21.
Kimura, T.D.1992"Hyperflow: A Visual Programming Language for Pen Computers" , Proceedings of the 1992 IEEE Workshop on Visual Lan guages, September 15-18, 1992, Seattle, Washington, pp 125-132
Kohonen, T.1988. "An Introduction to Neural Computing", Neural Networks, 1:3-16.
24.
Kohonen, T.1990. "The Self-Organizing Map", Proceedings of the IEEE, 78:9, 1464-1478.
25.
Lin, W.C.1982. "Computer Processing of Hand-Drawn Sketches and Diagrams", in Computer Analysis and Perception, Volume 1: Visual Signals, C. Y. Suen and R. De Mori, eds., Boca Raton, FL: CRC Press, Inc., pp. 107-149.
26.
Nellis, J. and T.J. Stonham.1992. "A Fully Integrated Hand-Printed Character Recognition System Using Artificial Neural Networks", Second IEE Conference ofArtificial Neural Networks, Bournemouth, U.K., IEE Publication No. 349, pp. 219-224.
27.
Reilly, D.L., L.N. Cooper and C. Elbaum.1982. 'A Neural Model for Category Learning", Biological Cybernetics, 45:35-41.
28.
Schwartz, J.1992. "The Revolution That Wasn't", Newsweek (October, 19):51.
29.
Servan-Schreiber, D., A. Cleeremans and J.L. McClelland.1988. "Encoding Sequential Structure in Simple Recurrent Networks", Carnegie Mellon University , November, 1988.
30.
Shynk, J.J. and S. Roy.1988"The LMS Algorithm with Momentum Updating", IEEE Symposium on Circuits and Systems, Espoo, Finland, NY: IEEE, pp. 2651-2654.
31.
Sorheim, E.1991. "ART2/BP Architecture for Adaptive Estimation of Dynamic Processes", Advances in Neural Information Processing Systems, Vol. 3, D. Touretzky, ed., pp. 169-175.
32.
Tappert, C.C., C.Y. Suen and T. Wakahara.1990. "The State of the Art in On-Line Handwriting Recognition" , IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(8):787-808.
33.
Taysi, B.M.1992. "Gesture System for a Graph Editor", master's thesis , Washington University in St. Louis, August, 1992.
34.
Thomas, Jr., G.B.1972. Calculus and Analytic GeometryReading, MA: Addison-Wesley Publishing Co., Inc. (cf. Thomas, 1972, 289-293 for energy integration calculation).
35.
Wagoner, J.1993. "Second Annual Reader's Choice Award Winners", Pen (March/April):11-21.
36.
Wang, C.1992. "A Fuzzy Approach to Handwritten Digit Recognition for Pen-Based Computers", master's thesis, Washington University in St. Louis, December 1992.
37.
Wang, C. , T. Kimura and T. Fuller1992. "Supervised Competitive Learning, Part II SCL with Fuzzy Logic", in Intelligent Engineering through Artificial Neural Networks , Volume 2, C. Dagh et al , ed., New York: ASME Press, pp. 351-356.
38.
Weiss, S.M. and C.A. Kulikowski.1991Computer Systems That LearnMorgan Kaufmann Publishers, Inc.
