Skip to content

Time-series forecasting with multiphoton quantum states and integrated photonics

Home / Papers / Time-series forecasting with multiphoton quantum states and integrated photonics

Time-series forecasting with multiphoton quantum states and integrated photonics

Authors: Rosario Di Bartolo, Simone Piacentini, Francesco Ceccarelli, Giacomo Corrielli, Roberto Osellame, Valeria Cimini, Fabio Sciarrino Date: 2026-04-17 Paper ID: openalex:2512.02928

Summary

This paper presents a quantum reservoir computing framework implemented on a reconfigurable integrated photonic circuit for time-series forecasting. The system encodes input signals into the circuit’s optical phases and employs a feedback mechanism modulated by measured output probabilities. A key finding is that input photon indistinguishability serves as a quantum resource, enabling the system to approximate higher-order nonlinear functions and outperform distinguishable input states in predictive tasks.

Key Contributions

  • Implemented a quantum reservoir computing protocol using reconfigurable integrated photonic circuits and single-photon detection.
  • Demonstrated that input photon indistinguishability acts as a resource for approximating higher-order nonlinear functions in time-series forecasting.
  • Showed that two-photon indistinguishable states achieve significantly superior forecasting performance compared to distinguishable input states.

Open Questions & Future Work

Key Concepts

  • photonic-quantum-reservoir-computing: A reservoir computing framework utilizing reconfigurable integrated photonic circuits and multiphoton quantum states for temporal data processing.

Archivist Review

The paper presents a specialized hardware-software co-design for reservoir computing. The concept of photonic quantum reservoir computing is approved as a foundational architectural category. The open question regarding the theoretical link between Fisher information and quantum computational expressivity is approved as it addresses a fundamental challenge in understanding why certain quantum states lead to superior nonlinear approximation capabilities.

Approved Concepts

  • Photonic Quantum Reservoir Computing: It is the core methodological contribution, combining integrated photonics with reservoir computing paradigms.

Approved Open Questions

  • Fisher information in QRC expressivity: Establishing this formal link would provide a theoretical foundation for optimizing photonic quantum reservoir designs beyond empirical trial-and-error, potentially identifying fundamental bounds on their expressive capacity.

Metadata & Links

url
https://arxiv.org/abs/2512.02928
paper_id
2512.02928
paper_source
openalex
domain
nlp
tags
architectures
datasets
concept_slugs
photonic-quantum-reservoir-computing
dataset_slugs
skill
TimeSeriesSkill
created_at
2026-04-20T05:54:18Z