[andrewjrangel]

Can someone help me Grok how the 3D printed Neural Networks back propagate? As I am trying to go through the paper they describe it as a pure optical approach, but what adjusts the refraction elements during the learning process?

[flohrian]

from the paper:

"[...] learnable network parameter that is iteratively adjusted during the training process of the diffractive network, using an error back-propagation method. After this numerical training phase implemented in a computer, the D^2NN design is fixed and the transmission/reflection coefficients of the neurons of all the layers are determined. This D^2NN design, once physically fabricated using e.g., 3D-printing, 3lithography, etc., can then perform, at the speed of light propagation, the specific task that it is trained for, using only optical diffraction and passive optical components/layers, creating an efficient and fast way of implementing machine learning tasks."

[andrewjrangel]

So the optical component is only the end result model of the NN? It isn't learning using the optics?

[dekhn]

Only indirectly. the physical device only does feedforward so they had to train it using tensorflow on a conventional device.

[cheeko1234]

Check out this new article:

https://www.osa.org/en-us/about_osa/newsroom/news_releases/2...

They implemented a back propagation algorithm using just optical.

[dqpb]

Which is unfortunate, since learning is the compute intensive part

[heyitsguay]

Can't access the article, but the abstract reads like this is inference only. If that's so, this still could be useful. For instance, you could train a network somewhere else, then implement the learned function optically.