Real space mapping of topological invariants using artificial neural networks
Authors:
Preprint
Abstract
Topological invariants allow to characterize Hamiltonians, predicting the existence of topologically protected in-gap modes. Those invariants are generically computed for infinite systems with trans- lational symmetry, by tracing the evolution of the occupied wavefunctions under twisted boundary conditions. However, those procedures do not allow to calculate a topological invariant by evaluating the system locally, nor allow to determine it in the absence of translational symmetry. Here we show that artificial neural networks can be trained to identify the topological order by evaluating a local projection of the density matrix. We demonstrate this for two different models, a 1-D topological su- perconductor and a 2-D quantum anomalous Hall state, both with spatially modulated parameters. Our neural network correctly identifies the different topological domains in real space, predicting the location of in-gap states. By combining a neural network with a calculation of the electronic states that uses the Kernel Polynomial Method, we show that the local evaluation of the invariant can be carried out for systems without translational symmetry consisting of tens of thousands of atoms. Our results show that supervised learning is an efficient methodology to characterize the local topology of a system.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- Learning phase transitions by confusionNature Physics, 2017
- Tunable symmetry breaking and helical edge transport in a graphene quantum spin Hall stateNature, 2013
- Mapping topological order in coordinate spacePhysical Review B, 2011
- Majorana fermions in semiconductor nanowiresPhysical Review B, 2011
- Computing topological invariants without inversion symmetryPhysical Review B, 2011
- Quantum anomalous Hall effect in graphene from Rashba and exchange effectsPhysical Review B, 2010
- Speech recognition with artificial neural networksDigital Signal Processing, 2010
- Intrinsic and Rashba spin-orbit interactions in graphene sheetsPhysical Review B, 2006
- Learning representations by back-propagating errorsNature, 1986
- The perceptron: A probabilistic model for information storage and organization in the brain.Psychological Review, 1958