Today I'm mostly working on 'nnirror', my art project about training neural networks to recognize themselves.
The ego network is trained using a generative adversarial network against the id network. Ego aims to recognize its own weights (output 1) vs everything else (output 0 for id's attempts to fool ego, output 0 for random input too).
The network weights are visualized at the top left of the first image, below is the normalized change since the previous epoch.
The second image plots the parameters (learning rates, momementa, etc), on the left if the ego network failed to achieve enlightenment after 1000 epochs, on the right if it managed to score above 4.5 in that time. The total score is twice the top graph minus the two lower graphs.
link to stroboscopic video Show more
I uploaded a short video of nnirror in action here (5mins, 60MB, no sound):
not sure about the border colour changing on the training view (it is redundant since I added the graphs), and the multicoloured lines on the other view for showing the learning parameters that led to success or failure are rather unilluminating...