Stochastic Convergence Analysis of the Single-Layer Backpropagation Algorithm for Noisy Input Data

IEEE Transactions on Signal Processing

Volumn 44, Issue 5, 1996, Pages 1315-1319

  Bershad, Neil J ^a   Cubaud, Nicolas ^a   Shynk, John J ^a  

^a University of California at Irvine   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

BACKPROPAGATION; CONVERGENCE OF NUMERICAL METHODS; IDENTIFICATION (CONTROL SYSTEMS); LEARNING ALGORITHMS; PROBABILITY; SIGNAL TO NOISE RATIO;

MEAN SQUARE ERROR (MSE); NOISY INPUT DATA; SINGLE LAYER BACKPROPAGATION ALGORITHM; STATISTICAL LEARNING BEHAVIOR; STOCHASTIC CONVERGENCE ANALYSIS;

SPURIOUS SIGNAL NOISE;

EID: 0030145768     PISSN: 1053587X     EISSN: None     Source Type: Journal    
DOI: 10.1109/78.502354     Document Type: Article

References (4)

1. B. Widrow and M. A. Lehr, "30 years of adaptive neural networks: Perccptron. madalinc and hackpropagation," Pmc. IEEE, vol. 78, pp. 1415-14.52, Sept. 1990.
2. N. J. Bershad, J. J. Shynk, and P. L. Feintuch, "Statistical analysis of the single-layer backpropagation algorithm: Part I-Mean weight behavior," IEEE 'Irans. Signal Processing, vol. 41, pp. 573-582, Feb. 1993.
3. "Statistical analysis of the single-layer backpropagation algorithm: Part II-MSB and classification performance," IEEE Trans. Signal Processing, vol. 41, pp. 583-591, Feb. 1993.
4. S. N. Diggavi. J. J. Shynk, and N. J. Bershad, "Convergence models for Rosenblatt's perceptron learning algorithm,'' IEEE Trans. Signal Processing, vol. 43, pp. 1696-1702, July 1995.