nips nips2000 nips2000-11 nips2000-11-reference knowledge-graph by maker-knowledge-mining

11 nips-2000-A Silicon Primitive for Competitive Learning


Source: pdf

Author: David Hsu, Miguel Figueroa, Chris Diorio

Abstract: Competitive learning is a technique for training classification and clustering networks. We have designed and fabricated an 11transistor primitive, that we term an automaximizing bump circuit, that implements competitive learning dynamics. The circuit performs a similarity computation, affords nonvolatile storage, and implements simultaneous local adaptation and computation. We show that our primitive is suitable for implementing competitive learning in VLSI, and demonstrate its effectiveness in a standard clustering task.


reference text

[1] M.A. Arbib (ed.), The Handbook of Brain Theory and Neural Networks, Cambridge, MA: The MIT Press, 1995.

[2] H.C. Card, D.K. McNeill, and C.R. Schneider,

[3] C. Diorio,

[4] P. Hasler,

[5] T. Allen et. aI,

[6] M. Lenzlinger and E.H. Snow,

[7] E. Takeda, C. Yang, and A. Miura-Hamada, Hot Carrier Effects in MOS Devices, San Diego, CA: Academic Press, 1995.

[8] T. Delbruck,

[9] D. Hsu, M. Figueroa, and C. Diorio,

[10] J. Lazzaro, S. Ryckebusch, M.A. Mahowald, and c.A. Mead,