A DYNAMIC K-WINNERS-TAKE-ALL NEURAL

In this paper, a dynamic K-winners-take-all (KWTA) neural network, whi ch can quickly identify the K-winning neurons whose activations are la rger than the remaining ones, is proposed and analyzed. For N competit ors, the proposed KWTA network is composed of N feedforward hardlimit neurons and three feedback neurons, which are used to determine the dy namic threshold. From theoretical analysis and simulation results, we found that the convergence of the proposed KWTA network, which require s Log,(N + 1) iterations in average to complete a KWTA process, is ind ependent of K, the number of the desired winners, and faster than that of the existing KWTA networks.