Error-structure-tailored early fault-tolerant quantum computing

TL;DR

This work develops an error-structure-tailored fault-tolerance framework that analyzes dissipative noise together with stabilizer-code structure to enable 1-fault-tolerant Z-rotations without T-gate magic state distillation. It introduces two schemes—expansion and projection—for preparing and injecting rotation states on surface codes, leveraging dispersive ZZ couplings and QED post-selection to achieve fault tolerance with small-angle errors scaling as O(|phi| p^2). The authors demonstrate 1-FT R_ZL(phi) gates on a [[4,1,1,2]] code and extend these ideas to surface codes, achieving rotation-state preparation with trace-distance scaling and favorable success probabilities, aided by Bayesian suppression and RUS-based injection. Through probabilistic coherent error cancellation and randomized control techniques, they quantify resource costs and show substantial spacetime-cost reductions relative to magic-state distillation and cultivation in Heisenberg-model simulations. The results indicate that millions of small-angle rotations are feasible at near-term hardware parameters, enabling practical early fault-tolerant quantum algorithms with reduced overhead and footprint.

Abstract

Fault tolerance is widely regarded as indispensable for achieving scalable and reliable quantum computing. However, the spacetime overhead required for fault-tolerant quantum computating remains prohibitively large. A critical challenge arises in many quantum algorithms with Clifford + $\varphi$ compiling, where logical rotation gates $R_{Z_L}(\varphi)$ serve as essential components. The Eastin-Knill theorem prevents their transversal implementation in quantum error correction codes and necessitating resource-intensive workarounds through T-gate compilation combined with magic state distillation and injection. In this work, we consider error-structure-tailored fault tolerance, where fault-tolerance conditions are analyzed by combining perturbative analysis of realistic dissipative noise processes with the structural properties of stabilizer codes. Based on this framework, we design 1-fault-tolerant continuous-angle rotation gates in stabilizer codes, implemented via dispersive-coupling Hamiltonians. Our approach could circumvent the need for T-gate compilation and distillation, offering a hardware-efficient solution that maintains simplicity, minimizes physical footprint, and requires only nearest-neighbor interactions. Integrating with recent small-angle-state preparation techniques, we can suppress the gate error to $91|\varphi| p^2$ for small rotation angle (where p denotes the physical error rate). For current achievable hardware parameters ($p=10^{-3}$), this enables reliable execution of over $10^7$ small-angle rotations when $|\varphi|\approx 10^{-3}$, meeting the requirements of many near-term quantum applications. Compared to the 15-to-1 magic state distillation and magic state cultivation approaches, our method reduces spacetime resource costs by factors of 1337.5 and 43.6, respectively, for a Heisenberg Hamiltonian simulation task under realistic hardware assumptions.

Figures (22)

• Figure 1: Illustration of the main idea. For the Clifford+$\varphi$ circuits, unlike the usual approach to compile the $R_Z(\varphi)$ gates to many $T$ gates and then perform magic state distillation or cultivation, we directly prepare the $\ket{r_\varphi}:=R_{Z}(\varphi)\ket{+}$ ancilla based on the error-structure-tailored FT analysis and perform injection.
• Figure 2: (a) $r$-filter is a projector $\Pi_{(r)}$ defined in \ref{['eq:Pir']} illustrated as a blue rectangle. Ideal decoder is illustrated as a green triangle followed by decoded logical block (black line). (b,c) The error-structure-tailored FT gadget requirements. We consider the noisy gate implementation as a dissipative process described by Lindblad master equation in \ref{['eq:Lindblad']}. We consider the Dyson expansion of the dynamics, and truncate to the $s$ order to consider the fault tolerance.
• Figure 3: (a) The $[[4,2,2]]$ code. The stabilizers are weight-4 $X$- and $Z$-operators on the data qubits. Treating one logical qubit as the gauge qubit, we obtain the $[[4,1,1,2]]$ subsystem code. (b,c) When the gauge qubit is fixed on the $Z$-basis (resp., $X$-basis), the $Z$-check (resp., $X$-check) can be written as two weight-2 gauge operators (shaded triangles), the code becomes the $[[4,1,2]]$ stabilizer code.
• Figure 4: Logical rotation gate $R_{Z_L}(\varphi)$ on the $[[4,1,1,2]]$ code by dispersively coupling the two data qubits $D_0$ and $D_2$ to a $3$-level g-f qubit. We assume the relaxation and dephasing error occuring on all the qubits during the whole procedure.
• Figure 5: Performance of the $1$-FT rotation gate on the $[[4,1,1,2]]$ code. We simulate the noisy $R_{Z_L}$ gate by preparing the logical rotation state $\ket{r_\varphi}_L$ by acting $R_{Z_L}$ on the $\ket{+}_L$ state. The performance is characterized by the trace distance to the ideal state. We consider both direct dispersive coupling (blue curves) and the ancillary-based approach (red curves) in \ref{['fig:RZZgate_ancilla']}.

Theorems & Definitions (14)

• Definition 1: $t$-FT gate gottesman2009intro
• Definition 2: Error-structure-tailored $t$-FT gate
• Proposition 1
• Proposition 2
• Proposition 3
• Definition 3: $t$-FT logical measurement for QED codes
• Definition 4: $t$-FT logical state preparation for QED codes
• Definition 5: $t$-FT gate for QED codes
• Definition 6: $t$-FT quantum error detection for QED codes
• Proposition 4