Perturbing the convolutional autoencoderΒΆ

So we have a convolutional autoencoder that works pretty well. Can we make it work even better - either improve the reconstruction accuracy or shrink the encoded state?

The all-convolutional architecture is nice, but the periodic boundary conditions are a nuisance to deal with. We can simplify the autoencoder by pre-scaling the data and rotating the interesting features away from the edges of the field.

To make the autoencoder more useful we can try two ways to shrink the encoded state: