Teaching a Neural Network Quantum Mechanics. A Deep Learning Approach to the NRepresentability Problem.
(2018) FYSM60 20181Mathematical Physics
Department of Physics
Solid State Physics
 Abstract
 We show that an unsupervised artificial neural network can be trained to parameterize the set of N representable density matrices well enough to enable ground state energy calculations. A onedimensional harmonic oscillator system is used to test the method. 4, 5, or 6 fermions are placed in an external potential. They interact with one of three different interaction types. By choosing the most successful network according to a welldefined measure, the approach is shown to generalize to interaction types not considered by the measure.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/studentpapers/record/8951887
 author
 Johansson, Emil ^{LU}
 supervisor

 MatsErik Pistol ^{LU}
 Gillis Carlsson ^{LU}
 Najmeh Abiri ^{LU}
 organization
 course
 FYSM60 20181
 year
 2018
 type
 H2  Master's Degree (Two Years)
 subject
 language
 English
 id
 8951887
 date added to LUP
 20180621 14:51:31
 date last changed
 20180621 14:51:31
@misc{8951887, abstract = {We show that an unsupervised artificial neural network can be trained to parameterize the set of N representable density matrices well enough to enable ground state energy calculations. A onedimensional harmonic oscillator system is used to test the method. 4, 5, or 6 fermions are placed in an external potential. They interact with one of three different interaction types. By choosing the most successful network according to a welldefined measure, the approach is shown to generalize to interaction types not considered by the measure.}, author = {Johansson, Emil}, language = {eng}, note = {Student Paper}, title = {Teaching a Neural Network Quantum Mechanics. A Deep Learning Approach to the NRepresentability Problem.}, year = {2018}, }