📄 tutor_nn_excute.m
字号:
% File Name : tutor_nn_excute.m
% Purpose : excuting the commands associated with the buttons of the
% RNN tutorial window
% Author : Hossam E. Mostafa Abdelbaki, School of Computer Science,
% University of Centeral Florida (UCF).
% Release : ver. 1.0.
% Date : October 1998.
%
% RNNSIM is a software program available to the user without any
% license or royalty fees. Permission is hereby granted to use, copy,
% modify, and distribute this software for any purpose. The Author
% and UCF give no warranty, express, implied, or statuary for the
% software including, without limitation, waranty of merchantibility
% and warranty of fitness for a particular purpose. The software
% provided hereunder is on an "as is" basis, and the Author and the
% UCF has no obligation to provide maintenance, support, updates,
% enhancements, or modifications.
%
% RNNSIM is available for any platform (UNIX, PCWIN, MACHITOCH).
% It runs under MATLAB ver. 5.0 or highrer.
%
% User feedback, bugs, or software and manual suggestions can
% be sent via electronic mail to : ahossam@cs.ucf.edu
function tutor_nn_excute(dir)
RnnTutorFigHndl = findobj('Tag','RnnTutorFig');
RnnTutorFigST1Hndl = findobj(RnnTutorFigHndl,'Tag','RnnTutorFigST1');
RnnTutorFigPrevPB = findobj(RnnTutorFigHndl,'Tag','RnnTutorFigPrevPB');
RnnTutorFigNextPB = findobj(RnnTutorFigHndl,'Tag','RnnTutorFigNextPB');
UD_PB1 = str2num(get(RnnTutorFigPrevPB,'UserData'));
set(RnnTutorFigST1Hndl,'FontSize',13 );
set(RnnTutorFigST1Hndl,'HorizontalAlignment','left');
switch dir
case 'Next'
UD_PB1 = UD_PB1 + 1;
case 'Prev'
UD_PB1 = UD_PB1 - 1;
case 'Quit'
close(RnnTutorFigHndl);
return
end
if (UD_PB1 < -1)
UD_PB1 = -1;
end
if(UD_PB1 >= 12)
UD_PB1 = 12;
end
qq = num2str(UD_PB1);
set(RnnTutorFigPrevPB,'UserData', qq);
if(UD_PB1 == -1 )
set(RnnTutorFigPrevPB, 'Enable', 'off');
set(RnnTutorFigNextPB, 'Enable', 'on');
tutorStr = ...
[' RNN TUTORIAL '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 0 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 1/11 '
' The RNN model allows arbitrary interconnections '
' between neurons. It has a mathematical proof that '
' it has a unique solution if the stability conditions are '
' met. By appropriately mapping external signals and '
' neuron states into certain physical quantities, it has '
' been applied successfully to several practical problems '
' The algorithm implemented in this program is based '
' on the following publication: '
' '
' E. Gelenbe "Learning in the recurrent random '
' network", Neural Computation, 5, pp 154-164, 1993. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 1 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 2/11 '
'Difference between the Random Neural Network (model/learning algorithm)'
'and other supervised Neural Network (models/learning algorithms): '
' ============================================================== '
'1- Its representation for the artificial neurons is more close to the '
' biophysical neurons. Where a neuron fires a train of impulses along '
' its axon when it is exited and after some time the neuron may fire '
' another train of pulses as a result of the same excitation. '
' (see Kandel, Schwartz, Principle of neural science , Elsevier, '
' Amsterdam, 1995. '
' '
'2- It has a general learning algorithm that can be applied to the feed '
' forward model ad the recurrent model. '
' '
'3- It needs more time during training than other models but once the '
' network is trained it becomes very fast (faster than the other '
' trained NN models) since there is no calculation of a non linear '
' function in the Random Neural Network ( RNN ) model. '];
set(RnnTutorFigST1Hndl,'Fontsize',10);
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 2 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 3/11 '
'4- The trained network has simple hardware implementation. '
' '
'5- All the weights resulting from training of the RNN are unsigned '
' numbers and this eliminates problems in the hardware implementation '
' of negative weights. Most NN models do not put restriction on the '
' sign or value of the weights. This limitation is hardly noticeable in'
' computer simulation but it becomes a critical issue when it comes to '
' dedicated hardware implementation, be they digital or analog. '
' (see: P.H. Graf and L.D. Jackel., "Analog electronic neural network '
' circuits," IEEE circuits and Devices magazine, pp. 44-55, July 1989 '
' T. Kohonen, Self Organization and Associative Memory. Springer Verlag,'
' Berlin, 1989.) '];
set(RnnTutorFigST1Hndl,'Fontsize',10);
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 3 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 4/11 '
' You can learn more about the RNN from the '
' following puplications : '
' ============================ '
' A- THE BASIC RNN THEORY '
' ============================ '
' 1- E. Gelenbe, "Random neural networks with '
' positive and negative signals and product '
' form solution," Neural Computation, Vol. 1, '
' No. 4, pp 502-510, 1989. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 4)
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 5/11 '
'2- E. Gelenbe, "Stability of the random neural network '
' model," Neural Computation, Vol. 2, No. 2, '
' pp 239-247, 1990. '
'3- E. Gelenbe, A. Stafylopatis, "Global behavior of '
' homogenous random neural systems," Applied '
' Mathematical Modelling, Vol. 15, pp 534-541, 1991. '
'4- E. Gelenbe "Learning in the recurrent random '
' network", Neural Computation, 5, pp 154-164, 1993. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 5)
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 6/11 '
' ================================ '
' B- APPLICATIONS IN OPTIMIZTION '
' ================================ '
'1- E. Gelenbe, V. Koubi, and F. Pekergin "Dynamical '
' random neural network approach to the traveling '
' salesman problem," ELEKTRIK, Vol. 2, No. 2, '
' pp l-10, 1994. '
'2- A. Ghanwani, "A qualitative comparison of neural '
' network models applied to the vertex covering '
' problem, ELECTRIK, Vol. 2, No. 2, pp 1l-18, 1994. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 6 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 7/11 '
'3- A. Ghanwani, "Neural networks for network optimization '
' Master thesis, Duke University, 1995. '
'4- E. Gelenbe "Learning in the recurrent random network", '
' Neural Computation, 5, pp 154-164, 1993. '
' ======================== '
' C- TEXTURE GENERATION '
' ======================== '
'1- V. Atalay, E. Gelenbe, and N. Yalabik "The random '
' neural network model for texture generation," '
' International Journal of Pattern Recognition and '
' Artificial Intelligence, Vol. 6(1),pp 131-141, 1992'];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 7 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 8/11 '
'2- V. Atalay, and E. Gelenbe "Parallel algorithm for '
' colour texture generation using the random neural '
' network model." International Journal of Pattern '
' Recognition and Artificial and-telligence, Vol. 6, '
' No. 2 & 3, pp 437-446, 1992. '
' ============================ '
' C- FUNCTION APPROXIMATION '
' =========================== '
'1- E. Gelenb, Z. H. Mao, and Y. D. Li, "Function '
' approximation with random neural network," to appear '
' in IEEE Trans. on Neural Networks Nov. 1998. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 8 )
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 9/11 '
' ============================== '
' D- IMAGE AND VIDEO COMPRESSION '
' ============================== '
'1- E. Gelenbe, C. Cramer, and M. Sungur,"Random '
' neural network learning and image compression," '
' Proceedings of the IEEE International Conference '
' on Neural Networks, pp. 3996-3999, 1994 '
'2- E. Gelenbe, C. Cramer, and M. Sungur, P. Gelenbe '
' "Traffic and video quality in adaptive neural '
' compression," Multimedia Systems, Vol. 4, pp. '
' 357-369, 1996. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 9)
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','on');
tutorStr = ...
[' 10/11 '
'3- C. Cramer, E. Gelenbe, and H. Bakircioglu "Low bit '
' rate video compression with neural networks and '
' temporal subsampling," Proceedings of the IEEE, '
' Vol. 84, No. 10, pp. 1529-1543, October 1996. '
'4- E. Gelenbe, T. Feng, and K.R.R. Krishnan "Neural '
' network methods for volumetric magnetic resonance '
' imaging of the human brain," Proceedings of the '
' IEEE, Vol. 84, No. 10, pp. 1488-1496, October 1996.'];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
if(UD_PB1 == 10)
set(RnnTutorFigPrevPB,'Enable','on');
set(RnnTutorFigNextPB,'Enable','off');
tutorStr = ...
[' 11/11 '
' ================= '
' E- BOOKS '
' ================= '
'1- E. Gelenbe (ed. and co-author), "Neural Networks: '
' Advances and Applications I, North-Holland Pub. '
' Co. (Amsterdam), 1991. '
'2- E. Gelenbe (ed. and co-author), "Neural Networks: '
' Advances and Applications II, North-Holland Pub. '
' Co. (Amsterdam), 1992. '];
set(RnnTutorFigST1Hndl,'String',tutorStr);
end
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -