Title :
Maximum entropy signal reconstruction with neural networks
Author :
Ingman, Dov ; Merlis, Yoram
Author_Institution :
Dept. of Quality Assurance & Reliability, Israel Inst. of Technol., Haifa, Israel
fDate :
3/1/1992 12:00:00 AM
Abstract :
The implementation of the maximum entropy reconstruction algorithms by means of neural networks is discussed. It is shown that the solutions of the maximum entropy problem correspond to the steady states of the appropriate Hopfield net. The choice of network parameters is discussed, and existence of the maximum entropy solution is proved
Keywords :
computerised signal processing; entropy; neural nets; Hopfield net; maximum entropy signal reconstruction; network parameters; neural networks; signal processing; steady states; Biological neural networks; Computer networks; Concurrent computing; Entropy; Filtration; Image processing; Neural networks; Parallel processing; Signal processing; Signal reconstruction;
Journal_Title :
Neural Networks, IEEE Transactions on
