How do neural networks work?

Watch the 3blue1brown series before the lecture: https://www.youtube.com/playlist?list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi

Moreover, read the Illustrated Guide to Recurrent Neural Networks by Michael Nguyen.

A quick recap

A simple NN. Source: towardsdatascience.com

Let’s review:

Different architectures

See the Neural Network Zoo by the Asimov Institute.

Example 1: FF for hand-written recognition

RNNs and LSTMS

An RNN

Example 2: RNNs and LSTMs

“Empirical ML”

Deciding the architecture

Activation functions

Choose a:

Read a simple explanation of activation functions here.

Loss functions

Choose:

Read this nice explanation on how to choose activation and loss functions.

Batch size and epochs

Dropout

Validation

Bibliography