There are numerous great tech blogs and discussions on neural network studies, just summarizing & understanding them will be good enough.

Blog
    FAQ on neural net.
    Overview of gradient optimization algorithms
    NN class (Stanford)



Discussion
    1. How to select the number of hidden layers and nodes?
    Summary: No hidden layer for linear separable questions, 1 hidden layer is usually enough for most, nodes # on hidden layer is mean of input & output for that layer.

• The number of hidden neurons should be between the size of the input layer and the size of the output layer.
• The number of hidden neurons should be 2/3 the size of the input layer, plus the size of the output layer.
• The number of hidden neurons should be less than twice the size of the input layer.