Researchers Measure Logical Qubit with 0.72% Error Using Quantum Surgery

Until now, achieving significant reductions in qubit overhead for fault-tolerant quantum computation required addressing limitations in selectively measuring individual logical qubits within encoded multi-qubit systems. Surabhi Luthra and Alex Owens demonstrated QLDPC surgery, a technique for performing non-destructive measurements on encoded quantum information, on a generalised bicycle code using Quantinuum’s H2-2 trapped ion processor. The team successfully demonstrated that QLDPC surgery performs operations on encoded quantum information without destroying it using Quantinuum’s trapped ion processor.

This innovation allows for more efficient use of qubits, the fundamental units of quantum computers, potentially lowering resource demands for building practical machines. Crucially, by reusing ancillary “gadget” qubits during measurements, they’ve taken an important step towards overcoming major obstacles to creating powerful and scalable quantum computing systems. The team and the National Quantum Computing Centre have achieved a key advance towards practical quantum computers by demonstrating QLDPC surgery on Quantinuum’s H2-2 trapped ion processor.

This is crucial because building powerful systems requires reducing the number of physical qubits needed to represent a single ‘logical qubit’, and certain error correction codes, known as QLDPC codes, offer significant potential here. Think of these codes like adding extra checks into a message so you can still understand it even if parts are lost or corrupted, but implementing them efficiently has been challenging until now.

QLDPC Surgery Demonstrates Subpoint Two Six Percent Logical Qubit

Error rates dropped to below 0.26% after implementing Quantum Low-Density Parity Check (QLDPC) surgery on Quantinuum’s H2-2 trapped ion processor. This result is a sharp improvement over prior lattice surgery demonstrations limited to single logical qubit blocks; those could not achieve such suppression through repeated measurements. The technique enables individual addressing of encoded qubits within multi-qubit systems, an important step towards reducing physical qubit overhead and realising order-of-magnitude reductions in resource demands for practical fault-tolerant quantum computation.

Specifically, the Quantinuum team successfully performed surgery on a multi-qubit system by measuring an individual logical qubit within a block encoding two qubits using QLDPC techniques. A single measurement utilising seven ancillary ‘gadget’ qubits achieved error rates of 0.72 [0.32, 1.41]% for one operator and 1.03 [0.60, 1.65]% for another; these results align with expectations from direct measurements of the encoded state.

Repeating this process three times to extract syndrome information, data revealing errors without collapsing the quantum state, reduced overall error below 0.26%, demonstrating fault tolerance; unaddressed “spectator” qubits remained stable during these rounds at only 0.21 [0.04, 0.67]% error. Further experiments reusing those same gadget qubits sequentially measured two different operators (X1 then Z2) with a combined error of 1.12 [0.60, 1.94]%. This reuse highlights an important efficiency gain in resource allocation.

Non-Destructive Syndrome Extraction Validates Core Principles for Scalable Quantum Error Correction

QLDPC surgery offers a pathway to drastically reducing the qubit resources needed for practical error correction; however, this demonstration focuses on validating the core “surgery primitive” itself, performing these delicate operations without destroying encoded information. While three rounds of syndrome extraction were achieved with diminishing errors, it remains unclear if that performance will hold as code blocks grow or more complex logical operations are layered upon one another. Acknowledging scaling challenges is sensible, and increasing block size or adding calculations will certainly present new hurdles in maintaining low error rates. This work confirms selective measurement of individual qubits within an encoded system using temporary “gadget” qubits, extra units solely used for measurement, and their reuse across multiple rounds minimizes resource demands while improving efficiency. By successfully performing non-destructive measurements on a generalised bicycle code utilising Quantinuum’s trapped ion processor, this approach validates key components needed for scalable quantum error correction schemes by demonstrating how syndrome information can be extracted without disturbing the quantum state itself.

This research demonstrated successful implementation of QLDPC surgery, a technique to selectively measure logical qubits, on a [[10,2,3]] generalized bicycle code with seven gadget qubits. The experiment achieved an error rate below 0.26% after three rounds of measurement and preserved unaddressed spectator qubits at less than 0.21% error, validating that encoded information remains stable during these operations. Reusing those same ancillary qubits to perform sequential measurements further improved resource efficiency; this work confirms principles for scalable quantum computation by showing how syndrome extraction can occur without collapsing the quantum state.

More like this

Quantum Error CorrectionResearchers Find Partial Decoding Boosts Quantum Error Correction

Quantum Error CorrectionResearchers Find ‘magic’ Fades Faster Than Entanglement

Quantum PhysicsBetter read-out boosts Gottesman-Kitaev-Preskill qubit performance

Quantum SensorsLogical qubits boost sensitivity to faint, random signals