Neural quantum states
Bohrdt Group · Research
Neural quantum states
Using restricted Boltzmann machines and other neural network architectures, we perform quantum state reconstruction from measurements and implement active learning to optimize measurement selection.
Neural quantum states compress the wave function into network parameters, enabling efficient simulation of strongly correlated systems beyond conventional methods. A recent advance is Gutzwiller-projected hidden fermion determinant states (G-HFDS) for the t-J model across the full doping range, competitive with MPS on large lattices at far fewer parameters.
Selected references
From architectures to applications: a review of neural quantum states
H. Lange, A. Van de Walle, A. Abedinnia, A. Bohrdt
Quantum Sci. Technol. 9, 040501 (2024) · DOI · arXiv:2402.09402
H. Lange, A. Van de Walle, A. Abedinnia, A. Bohrdt
Quantum Sci. Technol. 9, 040501 (2024) · DOI · arXiv:2402.09402
Simulating the two-dimensional t-J model at finite doping with neural quantum states
H. Lange, A. Böhler, C. Roth, A. Bohrdt
Phys. Rev. Lett. (2025)
H. Lange, A. Böhler, C. Roth, A. Bohrdt
Phys. Rev. Lett. (2025)
Neural network quantum states for the interacting Hofstadter model with higher local occupations and long-range interactions
F. Döschl, F. A. Palm, H. Lange, F. Grusdt, A. Bohrdt
Phys. Rev. B 111, 045408 (2025) · DOI
F. Döschl, F. A. Palm, H. Lange, F. Grusdt, A. Bohrdt
Phys. Rev. B 111, 045408 (2025) · DOI
Many-body dynamics with explicitly time-dependent neural quantum states
A. Van de Walle, M. Schmitt, A. Bohrdt
Mach. Learn.: Sci. Technol. 6, 045011 (2025) · DOI · arXiv:2412.11830
A. Van de Walle, M. Schmitt, A. Bohrdt
Mach. Learn.: Sci. Technol. 6, 045011 (2025) · DOI · arXiv:2412.11830
Neural quantum states for emitter dynamics in waveguide QED
T. Vovk, A. Van de Walle, H. Pichler, A. Bohrdt
DOI · arXiv:2508.08964
T. Vovk, A. Van de Walle, H. Pichler, A. Bohrdt
DOI · arXiv:2508.08964