Hedged maximum likelihood estimation
arXiv:1001.2029 · doi:10.1103/PhysRevLett.105.200504
Abstract
This paper proposes and analyzes a new method for quantum state estimation, called hedged maximum likelihood (HMLE). HMLE is a quantum version of Lidstone's Law, also known as the "add beta" rule. A straightforward modification of maximum likelihood estimation (MLE), it can be used as a plugin replacement for MLE. The HMLE estimate is a strictly positive density matrix, slightly less likely than the ML estimate, but with much better behavior for predictive tasks. Single-qubit numerics indicate that HMLE beats MLE, according to several metrics, for nearly all "true" states. For nearly-pure states, MLE does slightly better, but neither method is optimal.
4 pages + 2 short appendices
References in corpus (3)
Cited by in corpus (61)
- Universal Linear Optics
- Quantum Tomography via Compressed Sensing: Error Bounds, Sample Complexity, and Efficient Estimators
- Self-Consistent Quantum Process Tomography
- Maximum Likelihood, Minimum Effort
- Gate Set Tomography
- Quantum Machine Learning for Chemistry and Physics
- Quantum State Tomography via Linear Regression Estimation
- Adaptive quantum state tomography improves accuracy quadratically
- Quantum Algorithm Implementations for Beginners
- Detector Tomography on IBM 5-qubit Quantum Computers and Mitigation of Imperfect Measurement
- Permutationally invariant state reconstruction
- Systematic errors in current quantum state tomography tools
- Practical Bayesian Tomography
- Experimental progress on quantum coherence: detection, quantification, and manipulation
- Characterizing quantum dynamics with initial system-environment correlations
- Practical, Reliable Error Bars in Quantum Tomography
- Optimal two-qubit tomography based on local and global measurements: Maximal robustness against errors as described by condition numbers
- Optimal error regions for quantum state estimation
- Classical Shadows for Quantum Process Tomography on Near-term Quantum Computers
- Practical adaptive quantum tomography
- Projected Least-Squares Quantum Process Tomography
- Detecting correlated errors in SPAM tomography
- Rank-based model selection for multiple ions quantum tomography
- A comparative study of estimation methods in quantum tomography
- Constructing valid density matrices on an NMR quantum information processor via maximum likelihood estimation
- Adaptive quantum tomography
- Rank penalized estimation of a quantum system
- Errors in quantum tomography: diagnosing systematic versus statistical errors
- Priority Choice Experimental Two-qubit Tomography: Measuring One by One All Elements of Density Matrices
- A Practical Introduction to Benchmarking and Characterization of Quantum Computers
- Quantum Model Averaging
- Behavior of the maximum likelihood in quantum state tomography
- Simple, reliable and noise-resilient continuous-variable quantum state tomography with convex optimization
- Minimax quantum tomography: the ultimate bounds on accuracy
- Near-optimal quantum tomography: estimators and bounds
- Estimating the bias of a noisy coin
- Fourier Transform Quantum State Tomography
- Likelihood-free methods for quantum parameter estimation
- Evidence procedure for efficient quantum state tomography
- Inferring the Gibbs state of a small quantum system
- Maximum likelihood quantum state tomography is inadmissible
- Theory of versatile fidelity estimation with confidence
- Self-calibrating tomography for multi-dimensional systems
- Optimal measurement strategies for fast entanglement detection
- Maximum a posteriori estimation of quantum states
- Optimization of quantum state tomography in the presence of experimental constraints
- Pure state tomography with parallel unentangled measurements
- Efficient factored gradient descent algorithm for quantum state tomography
- Process estimation in presence of time-invariant memory effects
- Statistical analysis of low rank tomography with compressive random measurements
- Quantum Steering on IBMQ
- Entanglement characterization by single-photon counting with random noise
- Reliable Characterization for Improving and Validating Accurate Quantum Operations
- Minimax mean estimator for the trine
- Least-bias state estimation with incomplete unbiased measurements
- Experimentally undoing an unknown single-qubit unitary
- Reducing Complexity of Shadow Process Tomography with Generalized Measurements
- Quantum Tomography by Regularized Linear Regression
- Maximum expectation of observables with restricted purity states
- Convergence of the Exponentiated Gradient Method with Armijo Line Search
- Maximum-Likelihood Quantum State Tomography by Cover's Method with Non-Asymptotic Analysis