Abstract We study continuous-variable photonic quantum extreme learning machines as fast, low-overhead front-ends for collider data processing. Data is encoded in photonic modes through quadrature displacements and propagated through a fixed-time Gaussian quantum substrate. The final readout occurs through Gaussian-compatible measurements to produce a high-dimensional random feature map. Only a linear classifier is trained, using a single logistic regression, so retraining is fast, and the optical path and detector response set the analytical and inference latency. We evaluate this architecture on two representative classification tasks, top-jet tagging and Higgs-boson identification, with parameter-matched multi-layer perceptron (MLP) baselines. Using standard public datasets and identical train, validation, and test splits, the photonic Quantum Extreme Learning Machine (QELM) outperforms an MLP with two hidden units for all considered training sizes, and matches or exceeds an MLP with ten hidden units at large sample sizes, while training only the linear readout. These results indicate that Gaussian photonic extreme-learning machines can provide compact and expressive random features at fixed latency. The combination of deterministic timing, rapid retraining, low optical power, and room temperature operation makes photonic QELMs a credible building block for online data selection and even first-stage trigger integration at future collider experiments.
Abstract We study continuous-variable photonic quantum extreme learning machines as fast, low-overhead front-ends for collider data processing. Data is encoded in photonic modes through quadrature displacements and propagated through a fixed-time Gaussian quantum substrate. The final readout occurs through Gaussian-compatible measurements to produce a high-dimensional random feature map. Only a linear classifier is trained, using a single logistic regression, so retraining is fast, and the optical path and detector response set the analytical and inference latency. We evaluate this architecture on two representative classification tasks, top-jet tagging and Higgs-boson identification, with parameter-matched multi-layer perceptron (MLP) baselines. Using standard public datasets and identical train, validation, and test splits, the photonic Quantum Extreme Learning Machine (QELM) outperforms an MLP with two hidden units for all considered training sizes, and matches or exceeds an MLP with ten hidden units at large sample sizes, while training only the linear readout. These results indicate that Gaussian photonic extreme-learning machines can provide compact and expressive random features at fixed latency. The combination of deterministic timing, rapid retraining, low optical power, and room temperature operation makes photonic QELMs a credible building block for online data selection and even first-stage trigger integration at future collider experiments.
Quantum Machine Learning (QML) has emerged as a promising framework for exploring how quantum dynamics may enhance data processing tasks. Here we investigate Quantum Extreme Learning Machines (QELMs), a quantum analogue of classical Extreme Learning Machines in which training is restricted to the output layer. Our architecture combines dimensionality reduction (via PCA or Autoencoders), quantum state encoding, evolution under an XX Hamiltonian, and projective measurement to produce features for a classical single-layer classifier. By analyzing the classification accuracy as a function of evolution time, we observe a sharp transition between low- and high-accuracy regimes, followed by saturation. Remarkably, the saturated performance is comparable to that obtained using Haar-random unitaries that generate maximally complex dynamics, even though the XX model is integrable and local. Our results indicate that this increase in performance correlates with the onset of entanglement, which improves the embedding of classical data in Hilbert space and leads to more separable clusters in measurement probability space. Thus, moderate entanglement can contribute positively to the structure of the data representation, improving learnability without necessarily implying quantum computational advantage. For the image-classification tasks studied here, namely MNIST, Fashion-MNIST, and CIFAR-10, the relevant evolution time is consistent with information exchange over short distances and, within the explored system sizes, does not show evidence of scaling with the full system size. This suggests that QELM performance in this regime relies only on limited entanglement and remains compatible with efficient classical simulation. Our results clarify how local quantum dynamics and moderate quantum correlations are already sufficient to generate useful feature representations for learning.
Quantum extreme learning machines (QELMs) are unconventional computing architectures that bear remarkable promise in both classical and quantum machine-learning tasks, such as the estimation of quantum state properties. However, the probabilistic nature of quantum measurements demands extensive repetitions for training to precisely estimate expectation values, imposing stringent trade-offs among experimental resources, acquisition time, and signal-to-noise ratio, particularly for large datasets. Here, we introduce a paradigm shift by training the QELM exclusively with intense classical fields, namely twin beams generated by stimulated emission, while performing inference directly on previously unseen genuine quantum input states to predict their quantum properties. This strategy dramatically reduces acquisition times while substantially enhancing the signal-to-noise ratio. Using frequency-bin encoded biphoton states, implemented here for the first time in a quantum machine learning architecture, we demonstrate entanglement witnessing of two-qubit states with (93 +- 4)% accuracy, multi-dimensional entanglement detection, and learning of the Hamiltonian governing photon-pair generation with a fidelity of (96 +- 4)%. Our results open a new pathway toward faster and more robust training of photonic quantum extreme learning machines for quantum feature extraction.
Reservoir models offer a hardware-efficient learning paradigm for noisy intermediate-scale quantum devices by exploiting untrained quantum dynamics as a fixed feature map and restricting optimization to a simple classical readout layer. We propose a quantum extreme learning machine implemented using gaussian boson sampling and an encoding strategy that achieves high classification accuracy while reducing optical resource requirements. Classical inputs are jointly encoded in the squeezing parameters and in the interferometer unitary, enabling sampling-based, highly nonlinear feature maps while leveraging large-scale GBS output statistics, which are conjectured to be classically intractable. We systematically compare multiple families of quantum features accessible in the same setup and find that photon-number sampling probabilities provide the best performance, consistent with their higher effective feature dimensionality. Finally, we benchmark against classical nonlinear baselines and analyse robustness under noisy scenarios, showing competitive performance with fewer trainable parameters and indicating practical promise for near-term photonic implementations.
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 QELM'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 highlight the potential of time-delay photonic implementation as an experimentally accessible QELM platform.
Continuous-variable (CV) photonic quantum neural networks (QNNs) offer a promising pathway toward scalable quantum machine learning; however, their practical applicability is constrained by physical limitations such as Gaussian-only operations, limited measurement access, and readout overhead. To study these restrictions, we explore hybrid continuous-variable quantum neural network architectures for medical imaging tasks. We design physics-aware QNNs based on photonic circuits composed of displacements, phase rotations, squeezing, and beam splitters, integrated with classical preprocessing and task-specific output layers. Using continuous-variable Gaussian simulation, we study mode count, circuit depth, squeezing constraints, and quadrature-based readout strategies on learning and generalization. Benchmarking on MedMNIST datasets shows that for binary classification (PneumoniaMNIST), the proposed CV-QNN achieves an F1-score of 81.61% (AUROC = 85%), performing on par with its discrete-variable counterpart. For multi-class (PathMNIST), the CV architecture obtains an accuracy of 86.04%, though F1-score drops to 29.60%. The physics audit demonstrates orthogonality in state fidelity (F ≈ 1) across both datasets, confirming successful quantum feature mapping in phase-space. Ablation loops reveal the network exhibits noise resilience to electronic noise while maintaining stable multi-class F1-score, while also revealing the trade-off between quadrature information gain and signal clarity on homodyne and dual-homodyne detection. The proposed CV-QNN framework helps bridge quantum photonic and machine learning.
Introduction Accurate forecasting of solar power generation is essential for grid stability, yet it remains difficult because of the strong volatility and non-linear dynamics of solar resources. Classical machine learning models have improved forecasting performance, but they often fail to capture rapid changes in generation. Methods We propose a hybrid forecasting framework, Convolutional Quantum Extreme Learning Machines (C-QELM), which combines a fixed gate-based quantum feature map with local time-series patches and uses the resulting high-dimensional representations in classical forecasters. To evaluate the approach, we conducted a controlled synthetic benchmark comparing C-QELM with a global Standard QELM, a convolutional purely classical control, and Random Fourier Feature (RFF) maps. We also performed rolling one-step-ahead experiments on three solar datasets, comparing C-QELM with an unaugmented baseline and a dimension-matched classical convolutional random-feature control (CRC). Results In the synthetic benchmark, C-QELM substantially outperformed the global QELM and achieved the lowest mean error among the non-task-informed random-feature methods, although its confidence intervals overlapped with those of CRC and the best RFF configuration. Results on the solar datasets were similarly mixed, indicating that performance depends on both the dataset and the downstream architecture. Discussion Overall, C-QELM does not provide an unconditional performance advantage, but it may reveal a conditional benefit in highly dynamic regimes. Its small register size, low circuit depth, patch-wise reset, and lack of quantum-parameter training make it a promising architecture for future testing on NISQ platforms.