Exercise on sparse autoencoders

Exercise 1

This exercise is the first of several posts I am writing, for those who want a mathematical and hands-on introduction to deep neural networks.

1. Read the series of notes on the topic of “Sparse Autoencoder” in the UFLDL Tutorial.

2. Complete the exercise at the end of the series, using Python, not Matlab. Sai Ganesh and I have prepared the instructions for a Python-based exercise, together with the relevant data and source code template.

   • images.npy

   • sparseautoencoder.py

   • sparseautoencoder-exercise.pdf

3. Please email me if you would like the Python solution for the above exercise

Exercise on deep neural networks   Artificial general intelligence for the internet of things