Home > Publications database > Investigating Machine Learning methods to replace Hybrid Monte Carlo in simulations of Hubbard Model > print

001     906503
005     20230123101916.0
024 7 _ |a 2128/30825
|2 Handle
037 _ _ |a FZJ-2022-01482
041 _ _ |a English
100 1 _ |a John, Chelsea Maria
|0 P:(DE-Juel1)187395
|b 0
|e Corresponding author
245 _ _ |a Investigating Machine Learning methods to replace Hybrid Monte Carlo in simulations of Hubbard Model
|f - 2021-12-06
260 _ _ |c 2021
300 _ _ |a vi, 83
336 7 _ |a Output Types/Supervised Student Publication
|2 DataCite
336 7 _ |a Thesis
|0 2
|2 EndNote
336 7 _ |a MASTERSTHESIS
|2 BibTeX
336 7 _ |a masterThesis
|2 DRIVER
336 7 _ |a Master Thesis
|b master
|m master
|0 PUB:(DE-HGF)19
|s 1646316490_24387
|2 PUB:(DE-HGF)
336 7 _ |a SUPERVISED_STUDENT_PUBLICATION
|2 ORCID
502 _ _ |a Masterarbeit, Rheinische Friedrich-Wilhelms-Universität Bonn, 2021
|c Rheinische Friedrich-Wilhelms-Universität Bonn
|b Masterarbeit
|d 2021
520 _ _ |a The thesis research involves the application of machine learning (ML) to various parts of a Monte Carlo algorithm called Hybrid Monte Carlo (HMC–also referred to as Hamiltonian Monte Carlo), with the hopes that the neural network (NN), once properly trained, will speed up parts of the HMC algorithm. I implemented a NN that replaces the force calculations needed by HMC. The NN has been very successful for a large hyper-parameter space, and improves computational scaling from volume cube (N^3) scaling (w/o NN) to volume square (N^2) scaling (w/ NN), where volume here represents the total space-time dimension of the problem. The physics that motivates these calculations involves strongly correlated electrons described by the Hubbard model on two-dimensional lattices of various geometries. This model has broad applicability to solid-state and condensed matter systems. I have successfully applied my NN to hexagonal lattices (relevant for graphene), square lattices, and also more complicated lattices such as the kagome lattice (this exhibit topological behavior). In all cases I quantified the regions of parameter space where the NN adequately replaced the force calculations of HMC, thus providing improved scaling. For regions where the NN failed, I looked at alternative NN architectures (such as Bayesian NNs). I have also looked at the possibility of replacing the entire HMC algorithm (except for the Metropolis-Hastings step) with a modified NN leapfrog using unsupervised/reinforcement learning.
536 _ _ |a 5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)
|0 G:(DE-HGF)POF4-5111
|c POF4-511
|f POF IV
|x 0
856 4 _ |u https://juser.fz-juelich.de/record/906503/files/PhysicsMasterThesis.pdf
|y OpenAccess
909 C O |o oai:juser.fz-juelich.de:906503
|p openaire
|p open_access
|p VDB
|p driver
|p dnbdelivery
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 0
|6 P:(DE-Juel1)187395
913 1 _ |a DE-HGF
|b Key Technologies
|l Engineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action
|1 G:(DE-HGF)POF4-510
|0 G:(DE-HGF)POF4-511
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-500
|4 G:(DE-HGF)POF
|v Enabling Computational- & Data-Intensive Science and Engineering
|9 G:(DE-HGF)POF4-5111
|x 0
914 1 _ |y 2022
915 _ _ |a OpenAccess
|0 StatID:(DE-HGF)0510
|2 StatID
920 _ _ |l yes
920 1 _ |0 I:(DE-Juel1)JSC-20090406
|k JSC
|l Jülich Supercomputing Center
|x 0
920 1 _ |0 I:(DE-Juel1)IAS-4-20090406
|k IAS-4
|l Theorie der Starken Wechselwirkung
|x 1
980 1 _ |a FullTexts
980 _ _ |a master
980 _ _ |a VDB
980 _ _ |a UNRESTRICTED
980 _ _ |a I:(DE-Juel1)JSC-20090406
980 _ _ |a I:(DE-Juel1)IAS-4-20090406

LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21