arXiv:2609.18347 · Applied Physics Reviews
Optical Quantum Computing
Hamza Hasnaoui, Leonardo Limongi, Taira Giordani, Beatrice Polacchi, Alberto Quaranta, Martino Bernard, Fabio Sciarrino, Mirko Lobino
Abstract
Under the label of optical quantum computing, there are a variety of protocols and experiments that use the quantum properties of light to achieve a computational advantage over classical computing machines. In this review, we describe some of the main implementations, which differ in the type of encoding and in how the computation is performed, whether using gates or cluster states. For each protocol, we describe advantages and challenges with an overview of the experimental results obtained, summarized in tables. Because of the great relevance achieved in this field, there is a section dedicated to non-universal quantum computation with photons, where boson sampling, variational quantum eigensolvers, and quantum machine learning applications are described. The aim is to give the reader the broadest overview of the applications where photons and their quantum properties play a key role in computation.
AI Summary
This review surveys how photons can support universal and task-specific quantum computation, comparing discrete-variable (DV) and continuous-variable (CV) encodings, gate-based and measurement-based architectures, and their experimental implementations. It combines introductory derivations with chronological tables of resource-state generation, delegated computing, and photonic algorithm demonstrations; it does not present a new computational protocol or experimental result. The universal-computing discussion explains measurement-induced gates, cluster-state computation and adaptive feedforward, fusion-based architectures, and CV Gaussian and non-Gaussian resources. Quantitative benchmarks drawn from the literature include a $1/9$ success probability for a standard postselected optical CNOT implementation, CV linear clusters exceeding $10^6$ modes, and protocol-dependent theoretical photon-loss tolerances reaching $50\%$ in measurement-based schemes. Selected fusion-based proposals improve reported loss tolerance from $1.6\%$ to $10.4\%$. These numbers describe different architectures and assumptions, rather than directly comparable hardware thresholds. The review also covers boson sampling and quantum-advantage claims, variational algorithms, quantum machine learning, and privacy-preserving delegated computation. The central limitations identified are optical loss, probabilistic photon generation and entangling operations, source indistinguishability, integration and multiplexing overhead, and detector requirements. CV cluster generation is deterministic, but finite squeezing introduces noise, and universality and error correction require experimentally challenging non-Gaussian resources, including GKP states. Near-term non-universal devices avoid full fault-tolerance overhead but remain constrained by noise and circuit depth. The coverage is explicitly selective rather than exhaustive.
Error found: The displayed MZI has only two independent parameters modulo global phase and therefore cannot implement the arbitrary single-qubit operations claimed, although it can prepare arbitrary pure states from a fixed input. The DV teleportation derivation uses the wrong phase sign: projection onto the explicitly defined |±_φ⟩ produces e^(-iφ), requiring X^m H U_z(-φ)|ψ⟩ rather than the printed positive-angle rotation. The CV cluster-state definition uses edge weight 1/2, while its subsequent stabilizers and nullifiers assume edge weight 1.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Journal reference
- Appl. Phys. Rev. 13, 031306 (2026)
- DOI
- 10.1063/5.0308063