A Discrete-Time Recurrent Neural Network with One Neuron for k-Winners-Take-All Operation
In this paper, a discrete-time recurrent neural network with one neuron and global convergence is proposed for k-winners-take-all (kWTA) operation. Comparing with the existing kWTA networks, the proposed network has simpler structure with only one neuron. The global convergence of the network can be guaranteed for kWTA operation. Simulation results are provided to show that the outputs vector of the network is globally convergent to the solution of the kWTA operation.
KeywordsDiscrete-time recurrent neural network Global convergence k-winners-take-all operation
Unable to display preview. Download preview PDF.
- 2.Hertz, J., Krogh, A., Palmer, R.: Introduction to the Theory of Neural Computing. Addison-Wesley, Massachusetts (1991)Google Scholar
- 8.Maass, W.: Neural computation with winner-take-all as the only nonlinear operation. Advances in Neural Information Processing Systems 12, 293–299 (1999)Google Scholar
- 16.LaSalle, J.: The Stability of Dynamical Systems. Society for Industrial Mathematics (1976)Google Scholar