The Community for Technology Leaders
Green Image
<p><it>Abstract</it>—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 <math><tmath>${\rm O}({\rm n}^{2^{\rm s}/(2^{\rm s} - 1)})$</tmath></math> where s is the depth of cascaded sorting networks. Application of the WTA network to other problems such as nonbinary majority is also included.</p>
Quadratic perceptron, sorting, winner-take-all, nonbinary majority, complexity.
Ja-Ling Wu, Yuen-Hsien Tseng, "On a Constant-Time, Low-Complexity Winner-Take-All Neural Network", IEEE Transactions on Computers, vol. 44, no. , pp. 601-604, April 1995, doi:10.1109/12.376175
91 ms
(Ver 3.3 (11022016))