Convergence and stability of quantized Hopfield networks operating in a fully parallel mode. (English) Zbl 1208.68168
Summary: Hopfield neural networks (HNNs) have proven useful in solving optimization problems that require fast response times. However, the original analog model has an extremely high implementation complexity, making discrete implementations more suitable. Previous work has studied the convergence of discrete-time and quantized-neuron models but has limited the analysis to either two-state neurons or serial operation mode. Nevertheless, two-state neurons have poor performance, and serial operation modes lose fast convergence, which is characteristic of analog HNNs. This letter is the first in the field analyzing the convergence and stability of quantized Hopfield networks (QHNs)-with more than two states-operating in fully parallel mode. Moreover, this letter presents some further analysis on the energy minimization of this type of network. The main conclusion drawn is that QHNs operating in fully parallel mode always converge to a stable state or a cycle of length two and any stable state is a local minimum of the energy.
MSC:
| 68T05 | Learning and adaptive systems in artificial intelligence |
Keywords:
Hopfield neural networksReferences:
| [1] | DOI: 10.1109/TVT.2003.822000 · doi:10.1109/TVT.2003.822000 |
| [2] | DOI: 10.1023/B:AMAI.0000038311.03614.f0 · Zbl 1074.68048 · doi:10.1023/B:AMAI.0000038311.03614.f0 |
| [3] | DOI: 10.1109/18.21239 · Zbl 0666.94024 · doi:10.1109/18.21239 |
| [4] | DOI: 10.1109/LCOMM.2006.060770 · doi:10.1109/LCOMM.2006.060770 |
| [5] | DOI: 10.1109/TVT.2008.919619 · doi:10.1109/TVT.2008.919619 |
| [6] | DOI: 10.1109/81.153645 · Zbl 0775.92005 · doi:10.1109/81.153645 |
| [7] | Haykin S., Neural networks: A comprehensive foundation (1999) · Zbl 0934.68076 |
| [8] | DOI: 10.1073/pnas.81.10.3088 · Zbl 1371.92015 · doi:10.1073/pnas.81.10.3088 |
| [9] | Hopfield J. J., Biological Cybernetics 52 pp 141– (1985) |
| [10] | DOI: 10.1016/S0925-2312(01)00337-X · Zbl 1016.68076 · doi:10.1016/S0925-2312(01)00337-X |
| [11] | DOI: 10.1109/25.892541 · doi:10.1109/25.892541 |
| [12] | DOI: 10.1109/IJCNN.2005.1555952 · doi:10.1109/IJCNN.2005.1555952 |
| [13] | DOI: 10.1002/(SICI)1520-684X(19990615)30:6<1::AID-SCJ1>3.0.CO;2-2 · doi:10.1002/(SICI)1520-684X(19990615)30:6<1::AID-SCJ1>3.0.CO;2-2 |
| [14] | DOI: 10.1109/IJCNN.1999.831560 · doi:10.1109/IJCNN.1999.831560 |
| [15] | Moon T. K., Mathematical methods and algorithms for signal processing (2000) |
| [16] | DOI: 10.1007/BF00363956 · Zbl 0637.65062 · doi:10.1007/BF00363956 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
