arXiv subjects

Marco Grillo

Publications and source records attributed to Marco Grillo.

3 recordsLinked to original sources

arXiv2026

Bayesian Deck-of-cards-based Ordinal Regression with Sequential Preference Elicitation

The Deck-of-cards-based Ordinal Regression (DOR) infers a value function from a ranking of reference alternatives in which the Decision Maker (DM) inserts blank cards between consecutive levels to express preference intensity. DOR, and its stochastic extension (SMAA-DOR), treat these answers as hard constraints defining a set of compatible value functions. We propose B-DOR, a probabilistic reformulation of DOR in which each pair of adjacent levels yields an ordinal observation, the declared direction and the number of cards, modelled through a cumulative-link likelihood that relates the number of blank cards to the latent value difference between alternatives. Two Bayesian inference algorithms are proposed: BAYES-DOR samples the whole posterior distribution by Hamiltonian Monte Carlo; FTRL-DOR tracks the maximum a posteriori estimate by constrained convex optimization. Moreover, through a multi-step elicitation process, elicitation can be spread over several short sessions reducing the cognitive burden on the DM. Both algorithms enjoy logarithmic regret bounds for prediction that hold for any sequence of DM responses and that guide the choice of the prior hyperparameters. A Monte Carlo study over 768 configurations shows that accuracy grows with the number of sessions, that blank cards add significant information over preference directions alone, that both algorithms maintain good performance under inconsistent answers, and that both outperform DOR and SMAA-DOR. An illustrative application to Italian regional healthcare performance demonstrates the practical applicability of the approach for building composite indicators.

stat.ML↗

arXiv2026

Towards Quantum Networks: Characterizing Raman Noise over Metropolitan-scale Fiber Network

The coexistence of quantum and classical signals in optical fiber infrastructures represents a major challenge for large-scale quantum networks, as noise sources such as Raman scattering can significantly impact entanglement distribution, and the quantum protocols based on it. In this work, we analyze Raman scattering in the C-band, used for entanglement distribution, generated by a classical the O-band signal. The main contribution of this study lays in investigating these effects in a real metropolitan-fiber network, moving beyond controlled laboratory experiments to deployed telecommunication environments. Measurements are performed over a 7 km metropolitan fiber link using commercial sources and narrowband lasers. The experimental results shows good agreement between the measurements taken under laboratory conditions, although, within the metropolitan-scale loop, localized spectral anomalies are observed in the deployed fibers. Therefore, our results show that, whenever a quantum signal propagates in the C-band alongside an O-band classical channel within the same fiber, careful selection of the operating frequency is required, as Raman scattering and other real-world noise sources can significantly affect the quality and stability of the quantum transmission. In particular, we identify spectral regions that are less affected by Raman noise, thereby providing practical guidelines for optimal quantum channel allocation. We demonstrate that Raman-induced noise constitutes a dominant contribution to the quantum signal-to-noise ratio (SNR) in realistic deployments, beyond background and detector noise. Overall, our findings offer practical insights for deploying quantum communication systems over existing fiber networks, supporting the development of robust and scalable quantum infrastructures.

quant-ph↗

arXiv2026

Optimization of C-band quantum traffic coexisting with O-band classical traffic: preliminary results

The coexistence of quantum and classical signals in the same optical fiber is a critical challenge for the deployment of quantum networks. Indeed, selecting an optimal channel for quantum signal transmission is crucial to minimize noise arising from co-propagating classical signals. This work experimentally investigates spontaneous Raman scattering (SpRS), a major source of noise in signals transmitted along the same fiber. Unlike most previous studies relying on narrow-linewidth laboratory lasers or architectures based on spatial or temporal multiplexing of quantum and classical signals, we employ commercial SFP optical transceivers and standard single-core single-mode fiber for the transmission of quantum and classical signals in the same fiber, reflecting conditions typical of deployed urban fiber infrastructures. Building on these measurements, we derive a compact and predictive model that captures the Raman scattering profile, enabling accurate estimation of SpRS noise as a function of source power, wavelength, and fiber length. A key outcome of this work is that the proposed model is independent of the specific optical source used, demonstrating its generality and robustness. The model can therefore be used for the identification of optimal C-band channels for quantum signal allocation, namely those least affected by SpRS noise generated by co-propagating O-band classical traffic. These results pave the way for a parameter-robust description of Raman scattering applicable to diverse fiber-based systems.

quant-ph↗