self-learningMC

Implementation of "Self-learning Monte Carlo method", Liu et al (2017)

implemented models

• Ising model

  • original model $$H = -J \sum_{\langle i, j \rangle} \sigma_i \sigma_j$$
  • effective model $$H = - \sum_i h_i \sigma_i $$

• 4 spin interaction model

  • original model $$H = -J \sum_{\langle i, j \rangle} \sigma_i \sigma_j - K \sum_{ijkl\in \square} \sigma_i \sigma_j \sigma_k \sigma_l$$
  • effective model
    $$H_{\mathrm{trial}} = E_0- \tilde J_1 \sum_{\langle i, j \rangle_1} \sigma_i \sigma_j - \tilde J_2\sum_{\langle i, j \rangle_2} \sigma_i\sigma_j - \tilde J_3\sum_{\langle i, j \rangle_3}\sigma_i\sigma_j$$

    

implemented algorithms

• Metropolis algorithm
• Wolff algorithm