Photonic Time-Delayed Quantum Extreme Learning Machine
Abstract
We propose and numerically simulate a photonic quantum extreme learning machine (QELM) based on non-linear interactions in a time-delay non-linear interferometer with photon-number-resolving detection. We theoretically demonstrate the reservoir's computational capability by binary classification of the non-linear two moons benchmark. Analysis of the learned output weights and alternative quantum readouts shows that compact feature sets based on photon-number correlations preserve classification accuracy while potentially reducing measurement complexity. We further show that the QELM maintains its classification performance under the effect of losses by naturally redistributing the learned weights across a larger set of available features. These results demonstrate the resilience of the time-delay photonic implementation and highlight its potential as an experimentally accessible QELM platform.
Related
- Versatile Quantum Machine Learning with an Ultra-low Power Photonic Quantum Reservoir Computer
- Nonlinear polarization effects in optical fiber-based extreme learning machine
- Learning photonic quantum states
- Optical Speckle Patterns as Nonlinear Feature Maps for Photonic Machine Learning
- Optical free space extreme learning machine for the implementation of emergent complex systems
BibTeX
@misc{protsenko2026photonic,
title = {{Photonic Time-Delayed Quantum Extreme Learning Machine}},
author = {Ekaterina Protsenko and Caterina Vigliar and Francesco Da Ros},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.07416},
eprint = {2609.07416},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.07416},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Photonic Time-Delayed Quantum Extreme Learning Machine
AU - Ekaterina Protsenko
AU - Caterina Vigliar
AU - Francesco Da Ros
PY - 2026
DA - 2026/09/07
PB - arXiv
DO - 10.48550/arxiv.2609.07416
UR - https://arxiv.org/abs/2609.07416
AB - We propose and numerically simulate a photonic quantum extreme learning machine (QELM) based on non-linear interactions in a time-delay non-linear interferometer with photon-number-resolving detection. We theoretically demonstrate the reservoir's computational capability by binary classification of the non-linear two moons benchmark. Analysis of the learned output weights and alternative quantum readouts shows that compact feature sets based on photon-number correlations preserve classification accuracy while potentially reducing measurement complexity. We further show that the QELM maintains its classification performance under the effect of losses by naturally redistributing the learned weights across a larger set of available features. These results demonstrate the resilience of the time-delay photonic implementation and highlight its potential as an experimentally accessible QELM platform.
ER -