Title :
On a constant-time, low-complexity winner-take-all neural network
Author :
Tseng, Yuen-Hsien ; Wu, Ja-Ling
Author_Institution :
Republic of China Army Gen. Headquarters, Taoyun, Taiwan
fDate :
4/1/1995 12:00:00 AM
Abstract :
A nearly cost-optimal winner-take-all (WTA) neural network derived from a constant-time sorting network is presented. The resultant WTA network has connection complexity O(n(2s/(2s-1))) where s is the depth of cascaded sorting networks. Application of the WTA network to other problems such as nonbinary majority is also included
Keywords :
computational complexity; neural nets; connection complexity; constant-time; constant-time sorting network; low-complexity winner-take-all neural network; nonbinary majority; Binary trees; Circuits; Delay; Neural networks; Neurofeedback; Neurons; Output feedback; Polynomials; Sorting; Very large scale integration;
Journal_Title :
Computers, IEEE Transactions on
