I'm going to try to explain my problem... Basicaly, i have used the matrix notation to calculate the weight associated to the right patterns by using the following formulas : s=[s1, s2 ….. sn] t=[t1, t2 ….. tn]. Where "s" is the matrix containing right input values; Where "t" is the matrix containing target values; SxT=W Where "W" is the weight matrix. The test of the input values could be done by this formula xW=[y_in1, y_in2,… ,y_inn]->y Where "x" is the matrix containing the input values which need to be tested. I have tested the weight matrix with the same input patterns I used to calculate it, but there were some mistakes. Indeed the result matrix did not give the right results. Bellow I'm trying to explain some steps of my calculation... It is the table containing my right patterns x1 x2 x3 x4 t1 t2 1 0 0 1 0 1 0 1 1 1 1 1 1 1 1 0 0 0 1 0 1 0 1 0 I have found the following weight matrix, [11] [11] W=[21] [12] By multiplying "W" by the right patterns, I have found these values, firt pattern [2 3]->[1 1] second pattern [4 4]->[1 1] third pattern [4 3]->[1 1] fourth pattern [3 2]->[1 1] You can see that these target values are all wrong, they should be as follow [0 1] for the first pattern [1 1] for the second pattern [0 0] for the third pattern [1 0] for the fourth pattern If someone has an idea. Let me know cheers !!