Bounds in the Projective Unitary Group with Respect to Global Phase Invariant Metric

Basic Information

• Paper ID: 2510.09765
• Title: Bounds in the Projective Unitary Group with Respect to Global Phase Invariant Metric
• Authors: Bhanu Pratap Yadav, Mahdi Bayanifar, Olav Tirkkonen (Aalto University, Finland)
• Classification: quant-ph cs.IT math.IT
• Publication Date: October 10, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.09765

Abstract

This paper investigates the global phase invariant metric in the projective unitary group PUn, which is of significant importance in universal quantum computation. The authors compute the volume and measure of small metric balls in PUn and derive Gilbert-Varshamov and Hamming bounds. Furthermore, they provide upper and lower bounds on the kissing radius of codebooks in PUn as a function of minimum distance, and utilize the lower bound on kissing radius to establish a tight Hamming bound. The paper establishes bounds on the distortion-rate function for uniform source quantization on PUn, analyzes the minimum distances of codebooks including the projective Pauli group, Clifford group, and projective groups of diagonal gates in the Clifford hierarchy, and verifies theoretical results through simulations.

Research Background and Motivation

Problem Definition

In quantum computing, the design of quantum algorithms can be viewed as decomposing unitary matrices using a set of universal gates. Since the global phase of a quantum system does not affect measurable properties, gate approximation should be considered in the projective unitary group PUn rather than in the unitary group or special unitary group.

Research Significance

1. Quantum Computing Foundations: PUn consists of equivalence classes of n×n unitary operations differing by global phase, making the projective unitary group fundamental for constructing reliable quantum gates and implementing universal quantum computation.
2. Practical Application Needs: In quantum circuit optimization, parameters such as T-count and T-depth are critical, requiring accurate theoretical bounds to guide design.
3. Theoretical Gap: While the small ball volumes of unitary groups, Grassmannians, and Stiefel manifolds are well understood, PUn still lacks in-depth research in volume analysis and theoretical bounds.

Limitations of Existing Methods

• Traditional operator norms and trace distance are significantly affected by global phase in distance determination.
• Lack of systematic theoretical bounds for codebooks in PUn.
• Existing analysis of ball packing and covering problems primarily focuses on Euclidean spaces, with insufficient research on non-Euclidean geometries.

Core Contributions

1. Volume Computation: First computation of the volume of the projective unitary group PUn and the measure of small metric balls.
2. Theoretical Bounds: Derivation of Gilbert-Varshamov lower bounds and Hamming upper bounds in PUn.
3. Kissing Radius Analysis: Provision of upper and lower bounds on the kissing radius of codebooks and establishment of tight Hamming bounds.
4. Distortion-Rate Function: Establishment of distortion-rate function bounds for uniform source quantization on PUn.
5. Specific Codebook Analysis: Computation of minimum distances for projective Pauli group, Clifford group, and diagonal gate groups in the Clifford hierarchy.
6. Covering Radius: Provision of lower bounds on covering radius and expected covering radius for random codebooks.

Methodology Details

Task Definition

Investigation of coding theory problems in the projective unitary group PUn = {αU | U ∈ Un, |α| = 1} using the global phase invariant metric: $d(U,V) = \sqrt{1-\frac{1}{n}|\text{Tr}(U^H V)|}$

Core Theoretical Framework

1. Volume Computation

Theorem 1: The volume of PUn is $\text{Vol}(PU_n) = \frac{(2\pi)^{\frac{n(n+1)}{2}}}{2\pi\sqrt{n}\prod_{i=1}^n(i-1)!}$

Corollary 1: As R → 0, the measure of metric ball B(R) in PUn is $\mu_d(B(R)) = c_n R^D (1 + O(R^2))$ where $c_n = (2\pi)^{-\frac{(n-1)}{2}} \frac{n^{\frac{n^2}{2}}}{\Gamma(\frac{n^2-1}{2}+1)\prod_{i=1}^n(i-1)!}$, and D = n² - 1 is the dimension of PUn.

2. Kissing Radius Bounds

Theorem 2: For any code (|C|, δ) in PUn, the kissing radius ϱ satisfies: $\underline{\varrho} \leq \varrho \leq \overline{\varrho}$ where:

• $\underline{\varrho} = \sqrt{1-\frac{\sqrt{1-\delta^2}}{2}}$
• $\overline{\varrho} = \sqrt{1-\frac{\sqrt{1+(1-\delta^2)^2}}{2}}$

Technical Innovations

1. Geometric Analysis: Utilization of the quotient geometry Un/U1 structure to compute volume through free and proper group actions.
2. Geodesic Midpoints: Employment of Lie group geodesic descriptions to find geometric midpoints between two points.
3. Optimization Methods: Resolution of kissing radius bounds through constrained optimization problems.
4. Heisenberg-Weyl Basis: Analysis of Clifford group minimum distance using completeness of orthonormal bases.

Experimental Setup

Data Generation

• Generation of 10⁸ unitary matrices uniformly at random on the unitary group using Haar measure.
• Natural acquisition of Haar measure on PUn through quotient structure.
• Verification for different dimensions n = 2, 4, 8.

Evaluation Metrics

1. Minimum Distance: δ = min{d(Ci,Cj) : Ci,Cj ∈ C, i ≠ j}
2. Kissing Radius: ϱ = sup{R : BCi(R) ∩ BCj(R) = ∅, ∀i ≠ j}
3. Covering Radius: ρ = max{min d(Pi, U) : U ∈ PUn}
4. Distortion: D(C) = Emin d²(P,Q) : P ∈ C

Comparison Codebooks

1. Projective Pauli group P̃n
2. Projective Clifford group G̃n
3. Diagonal Clifford hierarchy D̃n,k
4. Semi-Clifford codebook C̃n,k
5. Clifford+T codebook C̃l2,3
6. Clifford+S codebook C̃l2,4

Experimental Results

Main Results

1. Minimum Distance Analysis

Proposition 1: Minimum distances of various codebooks are:

• Projective Pauli group: δp = 1
• Projective Clifford group: δc = √(1 - 1/√2) ≈ 0.644
• Diagonal Clifford hierarchy: δd = √(1 - cos(π/2^k))

2. Bound Verification

• Minimum distance of Pauli matrices lies between GV bound and Hamming bound, demonstrating optimality.
• Clifford group exceeds GV bound for m=1,2, but performance degrades with increasing dimension.
• Diagonal Clifford hierarchy systematically falls below GV bound.

3. Distortion Performance

• Semi-Clifford codebooks exhibit "floor effect" after k>4, with limited improvement in average distortion.
• Clifford+T and Clifford+S codebooks approach theoretical bounds.
• Superior performance to random codebooks for small l values, comparable performance for large l values.

Ablation Studies

Comparison of different hierarchy level k impacts reveals:

• Increasing hierarchy level alone cannot significantly improve performance.
• Products of multiple high-level elements achieve performance comparable to or better than random codebooks.

Numerical Verification

Figures 1-7 demonstrate consistency between theoretical results and simulations:

• Theoretical formulas for small ball measures precisely match simulations in small distance ranges.
• Kissing radius bounds effectively enclose simulated midpoint distances.
• Systematic codebooks outperform random codebook approximations for covering radius.

Related Work

Coding Theory Foundations

This paper extends coding theory on Grassmannian manifolds to the projective unitary group, building upon:

• Henkel et al.'s ball packing bounds on Grassmannian and Stiefel manifolds.
• Dai et al.'s work on quantization bounds on Grassmannian manifolds.
• Pitaval et al.'s density analysis of ball-embedded Stiefel and Grassmann codes.

Quantum Computing Applications

• Fowler's fault-tolerant quantum gate construction methods.
• Kliuchnikov et al.'s Clifford+T gate approximation algorithms.
• Selinger's efficient approximation methods for single-qubit gates.

Conclusions and Discussion

Main Conclusions

1. Establishment of a complete coding theory framework in PUn, including volume formulas, bounds, and performance analysis.
2. Kissing radius analysis provides tighter bounds than traditional Hamming bounds.
3. Systematic codebooks (such as Clifford+T) achieve near-optimal performance in practical applications.

Limitations

1. Semi-Clifford codebooks have structural constraints leading to "floor effect".
2. For large cardinality codebooks, certain bounds may not be sufficiently tight.
3. Numerical computation faces computational complexity challenges in high-dimensional cases.

Future Directions

1. Investigation of precise bounds for higher-dimensional PUn.
2. Development of optimized codebooks for specific quantum computing tasks.
3. Exploration of combinations of quantum error correction codes and projective unitary group encoding.

In-Depth Evaluation

Strengths

1. Theoretical Completeness: First establishment of a complete coding theory framework for PUn.
2. Mathematical Rigor: All theorems provided with rigorous mathematical proofs.
3. Practical Value: Results directly applicable to quantum circuit design and optimization.
4. Sufficient Verification: Theoretical results validated through large-scale numerical simulations.

Weaknesses

1. Computational Complexity: Computation of certain bounds may be infeasible in high-dimensional cases.
2. Application Scope: Primarily focused on single-qubit and low-dimensional cases.
3. Optimization Space: Performance of certain codebooks still has room for improvement.

Impact

This work provides theoretical foundations for codebook design in quantum computing, with significant implications for quantum algorithm optimization and fault-tolerant quantum computation. The methodology demonstrates good reproducibility, with code and data available for further research.

Applicable Scenarios

1. Quantum circuit synthesis and optimization.
2. Gate sequence design in fault-tolerant quantum computing.
3. Performance analysis of quantum approximation algorithms.
4. Quantum coding theory research.

References

The paper cites 36 relevant references spanning quantum computing, coding theory, differential geometry, and other fields, providing a solid theoretical foundation for the research.