1. C.M. Bishop, Mixture density networks, NCRG/94/004, available from http://www.ncrg.aston.ac.uk.
2. Neural Networks for Pattern Recognition;Bishop,1995
3. Training a feed-forward network by feeding gradients forward rather than by back-propagation of errors;Brouwer;Neurocomputing,1997
4. K. Hiraoka, T. Shigehara, H. Mizoguchi, T. Mishima, S. Yoshizawa, On the overfitting of the five-layered bottleneck network, Proceedings. ICONIP ’99. Sixth International Conference on Neural Information Processing, Vol. 1, Perth, Australia, 1999, pp. 234–239.
5. K. Hiraoka, S. Yoshizawa, Recalling of many-valued functions by successive iteration on bottleneck networks, ICONIP’98 Fifth International Conference on Neural Information Processing, Kitakyushu, Japan, 1998, pp. 2653–2660.