The error surface of the 2-2-1 XOR network: The finite stationary points

Neural Networks

Volumn 11, Issue 4, 1998, Pages 683-690

  Sprinkhuizen Kuyper, Ida G ^a   Boers, Egbert J W ^a  

^a LEIDEN UNIVERSITY   (Netherlands)

Author keywords

Error surface; Learning; Local minima; Neural network; Saddle points; XOR problem

Indexed keywords

COMPUTATIONAL COMPLEXITY; ERROR ANALYSIS; LEARNING ALGORITHMS; THEOREM PROVING; TRANSFER FUNCTIONS;

SIGMOID TRANSFER FUNCTIONS; TAYLOR SERIES EXPANSION; TRAINING ALGORITHMS;

NEURAL NETWORKS;

ALGORITHM; ARTICLE; ERROR; LEARNING; MATHEMATICAL ANALYSIS; NERVE CELL NETWORK; PRIORITY JOURNAL; PROBLEM SOLVING;

EID: 0032096333     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0893-6080(98)00014-8     Document Type: Article

References (6)

1. Hamey, L. G. C. (1995). Analysis of the error surface of the XOR network with two hidden nodes. Computing Report 95/167C, Department of Computing, Macquarie University, NSW, Australia.
2. Lisboa P.J.G., Perantonis S.J. Complete solution of the local minima in the XOR problem. Network. 2:1991;119-124.
3. Rumelhart, D. E., & McClelland, J. L. and the PDP Research Group (1986). Parallel Distributed Processing, Vol. 1. Cambridge, MA: MIT Press.
4. Sprinkhuizen-Kuyper, I. G., & Boers, E. J. W. (1995). The error surface of the XOR network with two hidden nodes: the finite stationary points. Technical Report 95-39, Dept. of Computer Science, Leiden University, The Netherlands.
5. Sprinkhuizen-Kuyper I. G., Boers E. J. W. The error surface of the simplest XOR network has only global minima. Neural Computation. 8:1996;1301-1320.
6. Sprinkhuizen-Kuyper, I. G., & Boers, E. J. W. (1996). The error surface of the XOR network with two hidden nodes: stationary points with infinite weights. Technical Report 96-10, Dept. of Computer Science, Leiden University, The Netherlands.