From Magic State Distillation to Dynamical Systems

TL;DR

This paper introduces a novel approach to analyze and simulate Magic State Distillation protocols using dynamical systems theory, enabling better understanding and optimization of quantum resource distillation processes.

Contribution

The authors propose mapping MSD protocols to dynamical systems, allowing analysis, simulation, and visualization, including for exotic and concatenated codes, which was not previously possible.

Findings

The $[[15, 1, 3]]$ code can distill states for the $ oot T$ gate.

The $[[5, 1, 3]]$ code can distill the $T$ magic state.

Concatenation of codes can generate diverse MSD protocols with reduced overhead.

Abstract

Magic State Distillation (MSD) has been a research focus for fault-tolerant quantum computing due to the need for non-Clifford resource in gaining quantum advantage. Although many of the MSD protocols so far are based on stabilizer codes with transversal $T$ gates, there exists quite several protocols that don't fall into this class. Here we propose a method to map MSD protocols to iterative dynamical systems under the framework of stabilizer reduction. With the proposed mapping, we are able to analyze the performance of MSD protocols using techniques from dynamical systems theory, easily simulate the distillation process of input states under arbitrary noise and visualize it using flow diagram. We apply our mapping to common MSD protocols for $\ket{T}$ state and find some interesting properties: The $[[15, 1, 3]]$ code may distill states corresponding to $\sqrt{T}$ gate and the $[[5, 1, 3]]$ code can distill the magic state corresponding to the $T$ gate. Besides, we examine the exotic MSD protocols that may distill into other magic states proposed in [Eur. Phys. J. D 70, 55 (2016)] and identify the condition for distillable magic states. We also study new MSD protocols generated by concatenating different codes and numerically demonstrate that concatenation can generate MSD protocols with various magic states. By concatenating efficient codes with exotic codes, we can reduce the overhead of the exotic MSD protocols. We believe our proposed method will be a useful tool for simulating and visualization MSD protocols for canonical MSD protocols on $\ket{T}$ as well as other unexplored MSD protocols for other states.