phase-of-matter-by-machine-learning

This project introduces unsupervised machine learning in the study of phase of matter in physics.

Symmetry broken phases

Potts models(clock models) in square lattice. n=2 corresponds to Ising model.
Monto Carlo method is used to simulate spin configurations at different temperatues.
High dimensional data of spin configurations are reduced by PCA method.
It is clear to see symmetric pattern after dimension reduction.
Theory of symmetry broken phase may be formulated by unsupervised machine learning.

In XY model, different phases are not marked by symmetry breaking

Traditional dimension reduction cannot tell KT phase transition
Autoencoder method are used, including cnn, variational autoencoder(va), a combination of cnn and va.
At this stage, cnn method wins. But it is still not good to reconstuct the right spin configuration with vortex.
Further exploration is badly required. Autoencoder of topological objects such as vortex is not a topic of attention for the deep learning community, but it seems not a trivial problem.

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
XY		XY
README.md		README.md
config_table_add2.txt		config_table_add2.txt
config_table_add3.txt		config_table_add3.txt
config_table_add4.txt		config_table_add4.txt
data_simulation.ipynb		data_simulation.ipynb
dimensional reduction.ipynb		dimensional reduction.ipynb
z3.png		z3.png
z4.png		z4.png