Let TOk represents the target output of k-th neuron lying on the output layer. The error in prediction at k-th output neuron corresponding to a particular training scenario can be calculated as follows: Ek = 1 2 (TOk − OOk) 2. (10.6) For a training scenario, the error in prediction considering all output neurons can be determined like the following: E = P k=1 1 2 (TOk − OOk) 2, (10.7) where P represents the number of output neurons. The total error in prediction for all output neurons can be determined considering all training scenarios as follows: Etot = L l=1 P k=1 1 2 (TOkl − OOkl) 2, (10.8) where L indicates the number of training scenarios.
https://essayhope.com/wp-content/uploads/2020/02/logo-1-300x75.png 0 0 pyta https://essayhope.com/wp-content/uploads/2020/02/logo-1-300x75.png pyta2021-12-15 21:25:262021-12-15 21:25:26Let TOk represents the target output of k-th neuron lying on the output layer