SijingDu

By using the neural network functions to add configuration dependence to the local fermionic tensors, we significantly improve the expressive power of fermionic tensor networks states in ground state problems of fermionic Hamiltonians, such as the 2D Fermi-Hubbard model.

We use fermionic tensor networks to encode the wavefunction sign structure, providing an alternative ansatz to the existing fermionic NQS based on mean-field wavefunctions.

Late advertisement of our new work on a novel hybrid method of combing fermionic tensor networks and neural networks for tackling fermionic ground state problems. arxiv.org/abs/2506.08329

Interference, which gives rise to the sign problem in quantum Monte Carlo, also leads to a significant increase in computational hardness in tensor networks. @quantumunivie @ChrisJielun @JiaqingJ go.aps.org/40Duuxt

We introduce a novel perspective to efficient tensor network computations for quantum systems that ensures strict variationality in ground state calculations and has the potential to represent volume-law entangled states. link.aps.org/doi/10.1103/Ph…

arxiv.org/abs/2407.11235 Garnet Chan's opinionated article on quantum chemistry and quantum advantage

@matthiasbal Hi Matthias, really nice perspective to interpret transformer model as vector-spin model! Just wondering except for mean-field apporixmatation of the partition function (free energy), whether other forms of approximate free energy are possible, like Bethe free energy in BP?

New post on approaching transformer modules from statistical mechanics. mcbal.github.io/post/transform… We implement a differentiable vector-spin model whose couplings act as learnable parameters and probe it with data to find a transformer-like attention response. 1/9

Our work on a ML-based algorithm for calculating ground-state properties of quantum many-body systems has just been published on Physical Review Research! journals.aps.org/prresearch/abs…