Benchmarking global symmetry in two-dimensional tensor network algorithms
Philipp Schmoll and Román Orús
Phys. Rev. B 102, 241101(R) (2020) - Published 1 December, 2020
We implement and benchmark tensor network algorithms with symmetry for systems in two spatial dimensions and in the thermodynamic limit. Specifically, we implement -invariant versions of the infinite projected entangled pair states and infinite projected entangled simplex states methods. Our implementation of symmetry follows the formalism based on fusion trees from Schmoll et al. [Ann. Phys. 419, 168232 (2020)]. In order to assess the utility of implementing symmetry, the algorithms are benchmarked for three models with different local spin: the spin-1 bilinear-biquadratic model on the square lattice, and the kagome Heisenberg antiferromagnets (KHAFs) for spin- and spin-2. We observe that the implementation of symmetry provides better energies in general than nonsymmetric simulations, with smooth scalings with respect to the number of parameters in the ansatz, and with the actual improvement depending on the specifics of the model. In particular, for the spin-2 KHAF model, our simulations are compatible with a quantum spin-liquid ground state.
