# File:Example.jpg

(See Figure 1). The final neural network model (comprising the final weights) is derived by training the network with the construction (or training) sample. The objective of training is to derive (optimal) weights such that the outputs (i.e. predicted values) of the neural network is as close as possible to the desired outputs (i.e. actual values) for the observations in the training sample. The back-propagation algorithm is the most commonly used training algorithm. Essentially, it computes the output for each observation in the training sample given the existing weights and calculates the error (i.e. difference between the predicted and actual values). The back-propagation algorithm then feeds back the error through the neural network and adjusts the weights to minimize the error. The training stops when the weights no longer change significantly and the error no longer decreases.

## File history

Click on a date/time to view the file as it appeared at that time.

Date/Time | Thumbnail | Dimensions | User | Comment | |
---|---|---|---|---|---|

current | 02:30, 31 May 2012 | 620 × 430 (46 KB) | Hamid Nasiri (Talk | contribs) | (See Figure 1). The final neural network model (comprising the final weights) is derived by training the network with the construction (or training) sample. The objective of training is to derive (optimal) weights such that the outputs (i.e. predicted val |

- You cannot overwrite this file.

## File usage

The following 2 pages link to this file: