[Learn-data-science] Deep Learning and the Schrodinger Equation

[David Livermore]

Hello fellow Data Scientists,


I came across this paper today and I thought it might be of interest.  It uses convolutional neural networks (a topic which is I think is outside the scope of our current course, but we can try to discuss in the future), however, I think it still may be of interest to some of you.  I've attached a link to the paper, as well as to an article that tries to explain convolutional neural networks in a bit of detail.


Essentially, the paper treats a 2D potential well as an image, and trains a neural network to determine the ground state energy of the system, completely by-passing any ab-initio approach for the calculation (they still needed to do some ab-initio work in order to train the network).


Deep Learning and the Schrodinger Equation:

https://arxiv.org/abs/1702.01361


An intuitive guide to Convolutional Neural Networks:

https://medium.freecodecamp.org/an-intuitive-guide-to-convolutional-neural-networks-260c2de0a050


Have a good week!


Cheers,

David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.triumf.ca/pipermail/learn-data-science/attachments/20181030/65423ffd/attachment.html>