Abstract | ||
---|---|---|
In this paper, a new neural-network model called WINSTRON and its novel array architecture are proposed. Based on a competitive learning algorithm that is originated from the coarse-fine competition, WINSTRON can identify the k larger elements or the k smaller ones in a data set. We will then prove that WINSTRON converges to the correct state in any situation. In addition, the convergence rates of WINSTRON for three special data distributions will be derived. In order to realize WINSTRON, its array architecture with low hardware complexity and high computing speed is also detailed. Finally, simulation results are included to demonstrate its effectiveness and its advantages over three existing networks |
Year | DOI | Venue |
---|---|---|
1998 | 10.1109/72.712163 | IEEE Transactions on Neural Networks |
Keywords | Field | DocType |
hardware,neural network,convergence rate,systolic array,artificial neural network,competitive learning,winner take all,neural networks,pattern recognition,indexing terms,artificial neural networks,analog circuits,neural network model,computer architecture,arithmetic,convergence,computational modeling | Convergence (routing),Architecture,Hardware complexity,Computer science,Network architecture,Neural net architecture,Artificial intelligence,Artificial neural network,Competitive learning algorithm,Machine learning | Journal |
Volume | Issue | ISSN |
9 | 5 | 1045-9227 |
Citations | PageRank | References |
32 | 1.63 | 11 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jui-Cheng Yen | 1 | 266 | 23.97 |
Jiun-in Guo | 2 | 503 | 79.34 |
Hun-Chen Chen | 3 | 108 | 11.20 |