Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.11767 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909958800408576 |
|---|---|
| author | Frohnert, Felix Koridon, Emiel Polla, Stefano |
| author_facet | Frohnert, Felix Koridon, Emiel Polla, Stefano |
| contents | We introduce an unsupervised machine-learning framework that discovers optimally compressed representations of quantum many-body ground states. Using an autoencoder neural network architecture on data from $L$-site Fermi-Hubbard models, we identify minimal latent spaces with a sharp reconstruction quality threshold at $L-1$ latent dimensions, matching the system's intrinsic degrees of freedom. We demonstrate the use of the trained decoder as a differentiable variational ansatz to minimize energy directly within the latent space. Crucially, this approach circumvents the $N$-representability problem, as the learned manifold implicitly restricts the optimization to physically valid quantum states. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11767 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Minimal Representations of Fermionic Ground States Frohnert, Felix Koridon, Emiel Polla, Stefano Quantum Physics Strongly Correlated Electrons Machine Learning We introduce an unsupervised machine-learning framework that discovers optimally compressed representations of quantum many-body ground states. Using an autoencoder neural network architecture on data from $L$-site Fermi-Hubbard models, we identify minimal latent spaces with a sharp reconstruction quality threshold at $L-1$ latent dimensions, matching the system's intrinsic degrees of freedom. We demonstrate the use of the trained decoder as a differentiable variational ansatz to minimize energy directly within the latent space. Crucially, this approach circumvents the $N$-representability problem, as the learned manifold implicitly restricts the optimization to physically valid quantum states. |
| title | Learning Minimal Representations of Fermionic Ground States |
| topic | Quantum Physics Strongly Correlated Electrons Machine Learning |
| url | https://arxiv.org/abs/2512.11767 |