Analog Clustering Circuit
This chapter describes the mapping of a clustering algorithm into analog circuits and the design of the constituent circuit blocks. The experimental characterization of the individual blocks and of the clustering system are described. The clustering algorithm implemented is based on the K-Means algorithm, but differs in that the magnitude of the updates is independent of the input.
KeywordsMemory Cell Tunneling Junction Centroid Location Gate Current Distance Current
Unable to display preview. Download preview PDF.
- Bridges S, Figueroa M, Hsu D, Diorio C (2005) A reconfigurable VLSI learning array. In: Proceedings of 31st European conference on solid-state circuits (ESSCIRC 2005). Grenoble, France, pp 117–120Google Scholar
- Delbruck T (1991) Bump circuits for computing similarity and dissimilarity of analog voltages. In: Proceedings of international joint conference on neural networks, July 8–12, Seattle, Washington, pp 475–479Google Scholar
- Figueroa M, Bridges S, Diorio C (2005) On-chip compensation of devicemismatch effects in analog VLSI neural networks. In: Saul L, Weiss Y, Bottou L (eds) Advances in neural information processing systems, vol 17. MIT Press, Cambridge, MA, pp 441–448Google Scholar
- Hsu D, Figueroa M, Diorio C (2001) A silicon primitive for competitive learning. Adv Neural Inf Process Syst 13:713–719Google Scholar
- Hsu D, Bridges S, Figueroa M, Diorio C (2003) Adaptive quantization and density estimation in silicon. Adv Neural Inf Process Syst 15:1107–1114Google Scholar