Skip to content

Latest commit

 

History

History
34 lines (22 loc) · 668 Bytes

README.md

File metadata and controls

34 lines (22 loc) · 668 Bytes

Quantum imaginary time evolution steered by reinforcement learning

This is the numerical code for the article:
Quantum imaginary time evolution steered by reinforcement learning

Requirements

with pip/conda install

pytorch
matplotlib
gym
numpy

Run Code

python run.py

References

If you find this work helps your research or use the code, please consider to cite our official reference. Thank you.

Results

Results for the TFIM.

TFIM

Results for the Maxcut

MAXCUT