Researchers stitch quantum codes together without extra qubits

Understand this faster with AI
Researchers have successfully measured 25,000 randomly selected logical operators using a new technique called the “code quilt,” demonstrating its functionality across a large and varied dataset. Stasiu Wolanski of Riverlane and University College London details this construction, which directly measures logical operations between quantum codes without requiring additional ancillary qubits, a significant advantage for scaling quantum computers. The work introduces a subsystem code construction and two protocol variants, one of which increases the physical qubit degree by at most two, offering a potentially practical implementation for near-term devices. Empirical analysis using bivariate bicycle and surface codes confirms the technique preserves code capacity distance across the extensive operator set.
Code Quilt Construction: Universal Logic Without Ancilla Qubits The code quilt construction directly measures a joint logical operator between quantum codes without requiring an additional ancillary system, a departure from many existing fault-tolerant measurement schemes. The construction’s circuits demand no physical qubits beyond those already allocated for idle syndrome extraction within the individual codeblocks, streamlining the quantum resource requirements for complex logical operations. A detailed structural analysis accompanies the construction, establishing several conditions under which the code-capacity distance, a measure of error correction capability, is maintained throughout the measurement process. Researchers developed two distinct protocol variants, each with unique trade-offs between performance and practicality. The first variant, while not fully fault-tolerant in the large-distance limit, offers a particularly efficient circuit implementation, increasing the physical qubit degree by a maximum of two. This limited increase in qubit demand positions it as a potentially viable option for near-term quantum devices where qubit resources are constrained. Theorem 7.2 establishes that for a homogeneous quilt, where constituent codes share identical properties, the operational distance is at least the distance of the original codes, provided certain conditions are met. This theorem provides a formal guarantee of error correction performance within the quilt construction. The technique’s viability extends to established quantum code types, as demonstrated through its application to bivariate bicycle and surface codes. These codes served as testbeds for empirical analysis, confirming the preservation of code-capacity distance across a substantial dataset. Bivariate Bicycle & Surface Codes in Code Quilt Measurements These quilts facilitate direct measurement of logical operators between quantum codes without introducing additional ancillary qubits, a key advantage over other approaches to quantum error correction. Code-capacity-distance preserving quilts were found for measuring 25,000 randomly selected logical operators. Achieving an operational distance equivalent to the original source codes is a critical requirement for these code quilts, and the researchers demonstrated this preservation through the examples generated for bivariate bicycle and surface code combinations. Visual representations of these quilts, showing stitching operators as dashed lines and Pauli type indicated by colour, are included in the paper. The simplicity of the code quilt approach is a deliberate design choice, aiming to reduce engineering complexity in the pursuit of practical quantum computing. “Theoretical simplicity is desirable in a field where the ultimate goal is to reduce engineering complexity,” the authors state, highlighting a focus on streamlining the path toward building functional quantum devices. This work introduces a technique for performing arbitrary logical Pauli product measurements between codes, and examines its performance on bivariate bicycle and surface codes, offering a potentially valuable tool for future quantum architectures.
Subsystem Code Design for Direct Pauli-Product Measurement This approach relies on creating a subsystem code with locally defined gauge operators, preserving the original structure of the constituent codes even as some elements are redefined as gauges. The resulting code quilt possesses low-density parity-check (LDPC) gauge generators by its very design, a characteristic that aids in efficient error correction. The researchers construct explicit circuits to perform a logical measurement using the quilt construction, which also require no physical qubits beyond the ancillas already in place to perform idle syndrome extraction on the codeblocks in question. The method defines a code quilt as a partially initialised code, acknowledging that the measured product’s value is unknown at the outset.
Gauge Generator Analysis & Code-Capacity Distance Preservation The code quilt construction, a new method for measuring logical operators between quantum code blocks, relies on identifying sufficient conditions to preserve a critical metric: code-capacity distance. Researchers detailed these conditions, establishing a baseline for evaluating the technique’s reliability in maintaining error correction capabilities as code complexity increases.
The team employed simulated annealing optimisation to identify code quilts capable of measuring logical operators, successfully testing 25,000 selected operations.
Optimizing Sparse Detector Bases via Simulated Annealing Simulated annealing was important in identifying code quilts capable of measuring logical operators, a demonstration of the technique’s scalability and effectiveness across a large dataset. Researchers also constructed sparse inherited generators under conditions relating to local check presentations, providing insight into the structural properties of these code quilts. While these results illuminate the underlying principles, the specific quilts used in the numerical experiments were not pre-defined but rather discovered through the simulated annealing process, highlighting the algorithm’s ability to navigate complex solution spaces. The source code for this optimisation is publicly available, allowing for further investigation and refinement. A key consideration during optimisation was balancing weight and congestion, particularly concerning measurement-consistency detectors. First Protocol: Subsystem Code Circuit Performance Evaluation Circuit-distance evidence supports the viability of a new approach to measuring logical operators between quantum codes. Researchers gauged-fixed a subsystem code into a stabiliser code, directly measuring its generators and preserving established code-capacity distance bounds. These generators originate as products of original checks within the system, with analytical results focused on code-capacity distance and extending to the gauge-fixed stabiliser protocol. The subsystem protocol, illustrated with a ZZ-type measurement across two codeblocks of a bivariate bicycle code, relies on distinguishing between untouched stabilisers and those that give rise to detectors. Untouched stabilisers, originating from original-code stabilisers, remain unchanged throughout the process. These operators, initially gauge operators, are upgraded to stabilisers through the gauge-fixing process. In the subsystem protocol, not all physical operators function as circuit detectors, necessitating categorization to identify those that do. “The generators are products of original checks,” the researchers state, highlighting the foundational basis of the measurement process.
The team reduced the search for optimal circuits to a shorter description using a map preserving spacelike logical effects, analytically bounding the distance of remaining timelike error directions based on circuit structure. Second Protocol: Stabilizer Code & Distance Preservation Representative searches for bivariate bicycle and surface codes demonstrate preservation of code-capacity distance, exceeding previously established sufficient conditions for fault-tolerance. These findings suggest that looser analytical conditions for maintaining quantum error correction properties may be achievable with further research.
The team constructed two distinct protocols for performing logical measurements based on this subsystem code, each with unique characteristics and trade-offs in implementation. The first protocol relies on directly measuring the anticommuting generators of the quilt subsystem code, then multiplying the outcomes to determine the central stabilisers. This approach preserves code-capacity distance, though circuit-level performance is also dependent on the schedule used for extracting these measurements. The observed performance parity with idle memory suggests the approach does not introduce significant additional errors during logical operations. Comparison to Generalized Surgery & QLDPC Approaches Existing approaches to performing logical operations on quantum codes often rely on substantial overhead, either in time or additional qubits. Generalized surgery, for example, requires ancilla systems to mediate interactions between code blocks, while high-rate quantum low-density parity-check (QLDPC) codes demand complex circuit implementations. This work presents an alternative, directly measuring logical operators between codeblocks without introducing new ancilla qubits beyond those already used for syndrome extraction.
The team’s method contrasts with techniques detailed by Yoder et al., which combine code automorphisms with intra- and inter-block Pauli-product measurements. The new construction, termed a “code quilt,” offers a different path to reducing overhead, particularly in space complexity. While some advancements have focused on minimizing ancilla system size within generalized surgery, this approach bypasses the need for bridging ancilla systems altogether. This is achieved through a direct measurement strategy that acknowledges the distinct identity of each codeblock, avoiding the treatment of multiple blocks as a single, larger code. Empirical analysis reveals a trade-off between circuit depth and complexity. The logical error probability is comparable to that of idle memory for operations involving the 144-qubit bivariate bicycle code and distance-12 rotated surface code at physical error rates between 1.5× 10–3 and 3× 10–3. Source: https://arxiv.org/abs/2611.26329v1 More like thisQuantum PhysicsA new circuit reverses the standard quantum Zeno probabilityQuantum Error CorrectionSimple error codes may be enough for long-distance quantum networksQuantum PhysicsChalmers, Stockholm & Gothenburg Researchers Probe Fermi Gas StatesQuantum PhysicsMotzkin chains show power-law entanglement, not the expected logarithmStay currentSee today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals. Tags:
Tags
Source Information
Discussion
0 professional contributions
Sign in to join this professional discussion.
Be the first to add a constructive contribution.
