Polynomial neural surrogates for designing photonic quantum experiments
Abstract
Physics simulators can support the discovery of quantum experiments by predicting the states generated by experimental configurations. When these simulators are computationally expensive, repeated simulator calls can limit the search for experiments that generate a desired quantum state. Here, we develop a physics-inspired polynomial neural surrogate for PyTheus, a graph-based quantum-optics simulator, to predict quantum states and use it to design quantum experiments. Its polynomial activations are motivated by the relation between graph perfect matchings and the resulting state amplitudes. We train separate surrogate models for four-, six-, and eight-photon systems and show that they achieve higher prediction accuracy with fewer trainable parameters than standard multilayer perceptrons. We then use the trained surrogates for inverse design of GHZ, W, and linear-cluster states. For the larger systems, the surrogates also enable faster inverse design than direct optimization with PyTheus. These results suggest that incorporating the underlying physics into neural surrogates can provide an efficient approach to quantum experiment design.
Related
- An Exact Polynomial Task-Risk Bridge for Time-Multiplexed Photonic Quantum Reservoirs
- Rank-Dependent Error Bounds and Near-Optimality in Quantum Control via Hierarchical Tucker Surrogates
- Neural Fourier Surrogates for Data Reuploading Quantum Neural Networks
- Benchmarking Quantum Convolutional Neural Networks and their classical surrogates on long-range strongly correlated models
- Learning photonic quantum states
BibTeX
@misc{chaurasiya2026polynomial,
title = {{Polynomial neural surrogates for designing photonic quantum experiments}},
author = {Rohit Chaurasiya and Xuemei Gu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.06032},
eprint = {2610.06032},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.06032},
}
RIS
TY - UNPB
TI - Polynomial neural surrogates for designing photonic quantum experiments
AU - Rohit Chaurasiya
AU - Xuemei Gu
PY - 2026
DA - 2026/10/05
PB - arXiv
DO - 10.48550/arxiv.2610.06032
UR - https://arxiv.org/abs/2610.06032
AB - Physics simulators can support the discovery of quantum experiments by predicting the states generated by experimental configurations. When these simulators are computationally expensive, repeated simulator calls can limit the search for experiments that generate a desired quantum state. Here, we develop a physics-inspired polynomial neural surrogate for PyTheus, a graph-based quantum-optics simulator, to predict quantum states and use it to design quantum experiments. Its polynomial activations are motivated by the relation between graph perfect matchings and the resulting state amplitudes. We train separate surrogate models for four-, six-, and eight-photon systems and show that they achieve higher prediction accuracy with fewer trainable parameters than standard multilayer perceptrons. We then use the trained surrogates for inverse design of GHZ, W, and linear-cluster states. For the larger systems, the surrogates also enable faster inverse design than direct optimization with PyTheus. These results suggest that incorporating the underlying physics into neural surrogates can provide an efficient approach to quantum experiment design.
ER -