In previous post we built a Neural Network Model and found the accuracy of the model.
In this post we will go further into the algorithm again and understand a simple concept of Local and Global Minima. This helps us build a neural network model which works best for us.
Local vs. Global Minimum
- The neural network might give different results with different start weights.
- The algorithm tries to find the local minima rather than global minima.
- There can be many local minima’s, which means there can be many solutions to neural network problem
- We need to perform the validation checks before choosing the final model.
Images above are visual representation of the local and global minima.
The next post is about Hidden layers and their Roles.
