Hi ITensor Users,
Happy New Year!
Recently, I implemented a strictly single-site DMRG with subspace expansion (DMRG3S) method, following PhysRevB.91.155115 and incorporating a randomized SVD scheme as suggested in this recent paper: [2403.00562] Comment on "Controlled Bond Expansion for Density Matrix Renormalization Group Ground State Search at Single-Site Costs" (Extended Version).
I have tested the implementation on a limited set of models, mainly the spinful Hubbard model, and found that it works quite well. The code is available here:
I would greatly appreciate it if you find this code interesting and are willing to test it on your own models or provide any feedback.
Best,
Zhaochen