Semi-Metal to Valence-Bond-Solid Phase Transition in the Honeycomb Lattice Using Machine Learning Techniques

Phase transitions in strongly correlated systems are a subject of extensive research. Determining critical temperature or critical coupling parameters responsible for these phase transitions at the thermodynamic limit is computationally demanding. Recently, machine learning techniques have been successfully applied to obtain critical parameters for finite systems. In this study, we will focus on constructing the phase diagram for the optical SSH model in a honeycomb lattice at the thermodynamic limit by employing various machine learning techniques including phase detection method (PD) and learning by confusion (LBC) on finite lattices, which vastly reduce computational costs compared to traditional finite size scaling methods that require multiple lattice sizes.