Implementing 3D Wolff update algorithm and other algorithms for 2d TFIM

jywu20 commented 2 years ago

If the standard way to map Z2 gauge field into 2d transverse field Ising model is to be implemented (to settle #2 ), 3D Wolff update algorithm is required since the path integral formulation of a 2d TFIM is a 3d classical Ising model. Also, methods like worldline QMC should also be implemented.

jywu20 commented 2 years ago

Algorithms to be implemented:

[x] 3D Wolff update algorithm
[ ] world line Monte Carlo algorithm for TFIM
[ ] direct simulation of Z2 gauge theory
[ ] benchmark

jywu20 commented 2 years ago

The 3D Wolff update algorithm now is implemented, which reuses code from src\pure-z2-mapped-to-spin-chain\anisotropic_ising.jl.

It should be noted that the test results of the 3D Wolff algorithm seem weird, including:

even within the FM phase, the fluctuation is still strong, as opposed to the case in 2D Wolff update
the phase boundary obtained by my version of 3D Wolff algorithm seems to deviate from the standard version
the FM phase's megnatization is closer to 0.5 instead of 1

More test is needed to find out what is going on here.

jywu20 / thermal-z2

Implementing 3D Wolff update algorithm and other algorithms for 2d TFIM #3