Source-linked AI summary
Measurement-based quantum computation
H. J. Briegel, D. E. Browne, W. Dür, R. Raussendorf, M. Van den Nest
TL;DR
Quantum computation still faces unresolved questions about scalable realization and the boundary of quantum computational power. This article surveys measurement-based quantum computation, emphasizing entangled resource states, adaptive measurements, fault tolerance, experimental implementations, and links to statistical mechanics. It highlights that topological protection yields fault-tolerant cluster-state schemes with thresholds up to 7.5 × 10−3, while large-scale realization and the quantum advantage over classical simulation remain unresolved.
Problem
Large-scale quantum computation remains difficult, and it is not known whether universal quantum computers preclude efficient classical simulation.
Method
The article surveys measurement-based quantum computation, focusing on entangled resource states, adaptive measurements, fault tolerance, experiments, and connections to other fields.
Results
Topological error correction combined with cluster-state computation yields a two-dimensional local architecture with a fault-tolerance threshold of 7.5 × 10−3.
Takeaways & Limitations
MQC provides a framework for relating computational power to entanglement and for pursuing scalable computation in noisy environments.
Takeaways & Limitations
Large-scale optical-lattice implementations still lack solved single-site addressing, and whether universal quantum computation is classically hard remains unknown.
Abstract
from arXiv · showhide
Quantum computation offers a promising new kind of information processing, where the non-classical features of quantum mechanics can be harnessed and exploited. A number of models of quantum computation exist, including the now well-studied quantum circuit model. Although these models have been shown to be formally equivalent, their underlying elementary concepts and the requirements for their practical realization can differ significantly. The new paradigm of measurement-based quantum computation, where the processing of quantum information takes place by rounds of simple measurements on qubits prepared in a highly entangled state, is particularly exciting in this regard. In this article we discuss a number of recent developments in measurement-based quantum computation in both fundamental and practical issues, in particular regarding the power of quantum computation, the protection against noise (fault tolerance) and steps toward experimental realization. Moreover, we highlight a number of surprising connections between this field and other branches of physics and mathematics.
1 Introduction
Measurement-based quantum computation (MQC) offers a distinct framework in which adaptive measurements process information on entangled resource states. The one-way model uses a universal 2D cluster state, shifting computational structure into measurement patterns and resource-state properties.
- 1 Introduction: Quantum computation remains challenging because scalable devices, the full range of applications, new quantum algorithms, and the classical–quantum computational-power boundary are not fully understood.
- 1 Introduction: MQC processes quantum information through sequences of adaptive measurements rather than coherent unitary evolutions.
- 1 Introduction: The one-way model prepares a 2D cluster state independently of the algorithm, then specifies the algorithm through the order and bases of individual qubit measurements.
- 1 Introduction: The one-way computer’s computational resource is entirely contained in its entangled cluster state, connecting computational power to properties of the resource state.
- 1 Introduction: MQC has developed into an interdisciplinary field linking entanglement theory, graph theory, topology, computational complexity, logic, and statistical physics.
2 Experimental proposals and achievements
Experimental proposals target one-way quantum computation with optical lattices, photons, and hybrid matter systems. These approaches exploit cluster-state structure to accommodate parallel operations, nondeterministic entangling gates, and modular architectures, while site addressing and photonic efficiency remain practical constraints.
- 2 Experimental proposals and achievements: Optical lattices can generate cluster states across many atoms through parallel state-dependent entangling operations, but individual-site addressing remains an implementation obstacle.
- 2 Experimental proposals and achievements: The one-way model is universal: despite random measurement outcomes, adaptive procedures realize any quantum computation deterministically up to local Pauli corrections.
- 2 Experimental proposals and achievements: In photonic implementations, nondeterministic entangling gates can create cluster states offline, after which one-way computation proceeds deterministically.
- 2 Experimental proposals and achievements: Linear-optical demonstrations have established key principles, but truly scalable schemes require photon generation and detection efficiencies beyond current experiments.
- 2 Experimental proposals and achievements: Hybrid matter-based schemes isolate individual qubits, reducing correlated error, while modularity facilitates scaling to many qubits.
- 2 Experimental proposals and achievements: Percolation can enable scalable computation with nondeterministic gates using simple non-switching optical circuits, relying on the uniformity of cluster or graph states.
3 Topological protection of information and fault-tolerant computation
Measurement-based quantum computation supports fault tolerance through error-correcting structures embedded in cluster states, with thresholds established for optical and nearest-neighbor settings. A three-dimensional cluster maps to a surface code evolving in time, enabling topologically protected encoded operations and a two-dimensional local variant.
- The threshold theorem permits arbitrarily large quantum computations with arbitrary accuracy when elementary error rates remain below a threshold.
- Fault-tolerant optical computation is possible with gate error rate 10−4 and photon loss rate 3 × 10−3.With photon creation and detection as the only imperfections, the threshold is ηS ηD > 2/3.
- A 3D cluster state combines one-way universality with topological error correction, directly building error correction into the cluster lattice for probabilistic gate errors.The reported threshold is 6.7×10−3 for a model including imperfect cluster-state preparation.
- Topological protection maps a 3D cluster state to a surface code propagating in time, where measurement choices modify the code surface and remove selected qubits.Measuring regions in the Z-basis creates a non-trivial topology in which fault-tolerant gates can be encoded.
- Electric and magnetic holes encode qubits, while strings of Pauli operators connect or loop around holes and are dragged as holes move.Moving holes and fusing them implements encoded unitary gates and measurements; a CNOT is realized by moving electric holes around a magnetic hole.
- A two-dimensional fault-tolerant cluster-state variant uses translation-invariant nearest-neighbor interactions and reaches a threshold of 7.5 × 10−3.The architecture is described as suitable for optical lattices, superconducting-qubit arrays, and ion traps.
4 Entanglement as a resource for computational power
MQC studies how entanglement in resource states determines computational power, using universality and classical simulatability as complementary perspectives. Recent results identify universal and non-universal resources, but the relationship between quantum and classical computational power remains unresolved.
- Universality: In MQC, single-qubit measurements cannot add entanglement, so the resource state's entanglement structure carries the computation's full power.The 2D-cluster state is the central example of a universal resource.
- Universality: Universal state-preparation resources must be maximally entangled with respect to every type of entanglement.This follows because any entanglement generated in the output must already be present in the resource.
- Universality: 1D-cluster, GHZ, W, Dicke, and certain strongly correlated one-dimensional spin-system states are not universal resources because at least one entanglement type is nonmaximal.High entanglement alone therefore does not establish universality.
- Universality: Graph states on several regular two-dimensional lattices remain universal, even with defects reaching about 40%, corresponding to the classical site-percolation threshold in two dimensions.The result includes triangular, hexagonal, and Kagome lattices.
- Universality: A weaker universality notion requires reproducing classical outputs of arbitrary gate-array computations rather than preparing arbitrary quantum states, allowing resources without all extremal 2D-cluster entanglement features.Such resources can be constructed using matrix-product-state or projected-entangled-pair descriptions.
- Classical simulation: Classical-simulation studies use entanglement-based techniques, including entanglement width, and efficiently simulate many states previously identified as non-universal.Non-universality does not by itself rule out usefulness for specific quantum tasks.
- Classical simulation: Whether universal quantum computation resists efficient classical simulation, and is therefore exponentially more powerful, remains unknown.The paper presents recent results as early progress rather than a resolution.
5 MQC and classical statistical mechanics
MQC is linked to classical statistical mechanics through a mapping between resource-state overlaps and spin-model partition functions. This connection transfers computational questions between the two fields, including efficient simulation and partition-function completeness.
- Mapping MQC to spin models: MQC resource states can be associated with classical spin systems such as the Ising and Potts models, connecting computational questions to statistical mechanics.The mapping uses the lattice or graph underlying the spin model to determine the entangled resource.
- Mapping MQC to spin models: The Ising partition function is identified with a quantum amplitude given by an overlap between an entangled resource state and a product state.The entangled state encodes interaction geometry, while the product state specifies interaction strengths, fields, and temperature.
- Mapping MQC to spin models: Computing the partition function corresponds to a measurement pattern on the associated resource state in MQC.Measurement outcomes occur with probabilities determined by the relevant overlaps.
- Computational consequences: Efficiently computable partition functions correspond to MQC models with no computational advantage over classical devices, and vice versa.For example, solvability of the two-dimensional Ising model without magnetic fields implies efficient simulation of MQC on the toric-code state.
- Computational consequences: Universality of the 2D-cluster state was used to show that the 2D Ising model with magnetic fields is complete for representing partition functions of arbitrary q-state spin models on arbitrary lattices.The representation occurs in a complex parameter regime.
- Broader connections: These connections provide a route for expressing statistical-physics problems in quantum-mechanical terms and may support quantum algorithms for problems in that area.The paper also notes connections between MQC and mathematical logic.
6 Outlook
The one-way model opens experimental avenues while leaving large-scale realization and theoretical understanding as open challenges. The paper concludes that MQC remains an attractive platform for investigating quantum computation and its broader connections.
- Experimental outlook: Large-scale laboratory quantum computation beyond proof-of-principle demonstrations remains a central experimental challenge.The paper identifies optical lattices as promising candidates for creating large cluster states efficiently.
- Experimental outlook: Single-site addressing in optical lattices remains unsolved, while scalable photonic implementations require more efficient single-photon sources.Hybrid one-way-computation proposals combine advantages of different physical implementations, but their long-term practicality is unresolved.
- Theoretical outlook: Theoretical priorities include fault-tolerant schemes, new quantum algorithms, universality, classical simulation, and understanding entanglement's role.Deeper universality results could help identify resource states tailored to specific physical systems.
- Conclusion: The one-way model remains an attractive alternative platform for experimental and theoretical investigations of quantum computation and its ramifications.This conclusion follows the paper's discussion of new experimental avenues and cross-disciplinary connections.