filippomb / variational-graph-auto-encoders-tensorflow-2-spektral-

Variational Graph Auto-encoders (Tensorflow 2 + Spektral)

This is a re-implementation of the Graph Auto-encoder and Variational Graph Auto-encoder presented here.

This implementation is based on Tensorflow 2 and Spektral. Compared to the original implementation, the code is more compact and the scripts are self-contained.

The results obtained are slightly different from those reported in the original paper (the AUC is higher). Also, there is not a clear improvement in using the variational version compared to the determistic one.

variational-graph-auto-encoders-tensorflow-2-spektral-'s Issues

Extra information about what is happening after the last GCN layer

Hi,

This looks really interesting! I was reading into the paper from Kipf but there only was a tensorflow v1 implementation so this is perfect for experimentation and understanding.

I do have one question, what is exactly happening in the following lines of code?

out = tf.matmul(z, tf.transpose(z))
A_rec = tf.keras.layers.Activation('sigmoid')(out)
out = tf.reshape(out, [-1])

It seems that first you multiply encoded embedding called 'z' with its transpose.
That would collect information from neighbouring nodes? (by looking at this stackoverflow example)

Also, my interest is using these techniques for further analysis of the embedded features, so I think that I need do something different then:

A_rec = tf.keras.layers.Activation('sigmoid')(out)
out = tf.reshape(out, [-1])

if link prediction is not my goal? For example, clustering, classification, etc of the graph or the nodes.

Kind regards and thanks for this work!

Stefan