| name | wigner-function-reconstruction |
| description | Wigner function reconstruction methodology for continuous-variable quantum system characterization.
Combines provably efficient regression for sparse states (binomial codes, cat states) with deep learning
for general states (GKP). Use when: (1) characterizing CV quantum systems, (2) reconstructing Wigner functions
from sparse measurements, (3) identifying error processes in QEC cycles, (4) phase-space tomography,
(5) reducing measurement overhead in quantum state characterization. Trigger words: wigner function,
phase-space reconstruction, CV quantum characterization, GKP state tomography, sparse phase-space data.
|
Wigner Function Reconstruction
Core Insight
Wigner function learning characterizes continuous-variable (CV) quantum systems by inferring continuous
phase-space functions from sparse pointwise measurements. Complexity scales logarithmically with Hilbert
space dimension for sparse states.
Two-Regime Approach
Sparse States (Binomial, Cat States)
For states with sparse Fock-space or coherent-state representations:
- Use provably efficient regression models
- Measurement complexity: O(log(d)) where d = effective Hilbert dimension
- No deep learning needed — compressed sensing suffices
General States (GKP, Unknown States)
For states without sparse structure:
- Train deep learning model on sparse measurements
- Model generalizes to arbitrary phase-space resolution
- Key: learns to identify dominant error processes with fewer measurements than conventional estimation
Experimental Validation
Validated on circuit-QED experimental data:
- Reconstructs Wigner functions across multiple QEC rounds
- Identifies dominant error processes with significantly fewer measurements
- Works on both simulated and real experimental data
Pipeline
- Data collection: Sparse pointwise measurements in phase space
- State classification: Determine if state has sparse structure
- Method selection:
- Sparse → compressed sensing regression (O(log d) measurements)
- General → trained deep learning model
- Reconstruction: Generate continuous Wigner function
- Error identification: Identify dominant error processes from reconstructed function
Scripts
See scripts/wigner_reconstruct.py for implementation.
References
- arXiv:2607.06232 - "Learning to Reconstruct Wigner Functions in Phase Space"
- GKP codes: Gottesman-Kitaev-Preskill error correction
- Circuit QED: superconducting qubit-cavity systems
Activation Keywords
- wigner function, phase-space reconstruction, CV quantum characterization
- GKP state tomography, sparse phase-space data
- quantum state reconstruction, continuous variable tomography