Quantum computing shows promise for addressing computationally intensive problems but is constrained by the exponential resource requirements of general quantum state tomography (QST), which fully characterizes quantum states through parameter estimation. We introduce the QST with Chebyshev polynomials, an approximate tomography method for pure quantum states encoding complex-valued functions. This method reformulates tomography as the estimation of Chebyshev expansion coefficients, expressed as inner products between the target quantum state and Chebyshev basis functions, measured using the Hadamard test circuit. By treating the truncation order of the Chebyshev polynomials as a controllable parameter, the method provides a practical balance between efficiency and accuracy. For quantum states encoding functions dominated by large-scale features, such as those representing fluid flow fields, appropriate truncation enables faithful reconstruction of the dominant components via quantum circuits with linear depth, while keeping both measurement repetitions and post-processing independent of qubit count, in contrast to the exponential scaling of full measurement-based QST methods. Validation on analytic functions and numerically generated flow-field data demonstrates accurate reconstruction and effective extraction of large-scale features, indicating the method's suitability for systems governed by macroscopic dynamics.
Quantum computing shows promise for addressing computationally intensive problems but is constrained by the exponential resource requirements of general quantum state tomography (QST), which fully characterizes quantum states through parameter estimation. We introduce the QST with Chebyshev polynomials, an approximate tomography method for pure quantum states encoding complex-valued functions. This method reformulates tomography as the estimation of Chebyshev expansion coefficients, expressed as inner products between the target quantum state and Chebyshev basis functions, measured using the Hadamard test circuit. By treating the truncation order of the Chebyshev polynomials as a controllable parameter, the method provides a practical balance between efficiency and accuracy. For quantum states encoding functions dominated by large-scale features, such as those representing fluid flow fields, appropriate truncation enables faithful reconstruction of the dominant components via quantum circuits with linear depth, while keeping both measurement repetitions and post-processing independent of qubit count, in contrast to the exponential scaling of full measurement-based QST methods. Validation on analytic functions and numerically generated flow-field data demonstrates accurate reconstruction and effective extraction of large-scale features, indicating the method's suitability for systems governed by macroscopic dynamics.
Abstract Quantum tomography is a crucial tool for characterizing quantum states and devices and estimating nonlinear properties of the systems. Performing full quantum state tomography on an N qubit system requires an exponentially increasing overhead with O(3^N) distinct Pauli measurement settings to resolve all complex phases and reconstruct the density matrix. However, many potential quantum computing applications, such as linear system solves, require only real-valued amplitudes. We introduce a readout method for real-valued quantum states that reduces measurement settings required for state vector reconstruction to O(N); the post-processing cost remains exponential Ω(2^N). This approach offers a substantial speedup over conventional tomography. We experimentally validate our method up to 10 qubits on the latest available IBM quantum processor and demonstrate that it accurately extracts key properties such as entanglement and magic. Our method also outperforms the standard SWAP test for state overlap estimation. This calculation resembles a numerical integration in certain cases and can be applied to extract nonlinear properties, which are important in application fields. We further implement the method to readout the solution from a quantum linear solver.
Quantum state tomography (QST) is a crucial tool for characterizing quantum states. However, performing full QST becomes impractical for reconstructing multiqubit density matrices since datasets and computational costs grow exponentially with qubit number. To bypass full QST, we introduce selective and efficient QST (SEEQST)—a method that enables efficient estimation of multiple selected elements of an arbitrary multiqubit density matrix. We show that any N -qubit density matrix can be partitioned into 2 N disjoint subsets, each containing 2 N elements. With SEEQST, any such subset can be accurately estimated from just two experiments with only single-qubit measurements. The complexity for estimating any subset remains constant regardless of Hilbert-space dimension, so, if desired, SEEQST can reconstruct the full density matrix, using 2 N + 1 − 1 experiments, where standard methods would use 3 N experiments. We provide a circuit decomposition for the SEEQST experiments, demonstrating that their maximum circuit depth scales logarithmically with N assuming all-to-all connectivity.
Abstract Quantum state tomography (QST) is a widely employed technique for characterizing the state of a quantum system. However, it is plagued by two fundamental challenges: computational and experimental complexity grows exponentially with the number of qubits, rendering experimental implementation and data post-processing arduous even for moderately sized systems. Here, we introduce gradient-descent (GD) algorithms for the post-processing step of QST in discrete- and continuous-variable systems. To ensure physically valid state reconstruction at each iteration step of the algorithm, we use various density-matrix parameterizations: Cholesky decomposition, Stiefel manifold, and projective normalization. These parameterizations have the added benefit of enabling a rank-controlled ansatz, which simplifies reconstruction when there is prior information about the system. We benchmark the performance of our GD-QST techniques against state-of-the-art methods, including constrained convex optimization, conditional generative adversarial networks, and iterative maximum likelihood estimation. Our comparison focuses on time complexity, iteration counts, data requirements, state rank, and robustness against noise. We find that rank-controlled ansatzes in our stochastic mini-batch GD-QST algorithms effectively handle noisy and incomplete data sets, yielding significantly higher reconstruction fidelity than other methods. Simulations achieving full-rank seven-qubit QST in under three minutes on a standard laptop, with 18 GB of RAM and no dedicated GPU, highlight that GD-QST is computationally more efficient and outperforms other techniques in most scenarios, offering a promising avenue for characterizing noisy intermediate-scale quantum devices. Our Python code for GD-QST algorithms is publicly available at github.com/mstorresh/GD-QST .
Quantum state tomography, the ability to deduce the density matrix of a quantum system from measured data, is of fundamental importance for the verification of present and future quantum devices. It has been realized in systems with few components but for larger systems it becomes rapidly infeasible because the number of quantum measurements and computational resources required to process them grow exponentially in the system size. Here we show that we can gain an exponential advantage over direct state tomography for quantum states typically realized in nature. Based on singular value thresholding and matrix product state methods we introduce a state reconstruction scheme that relies only on a linear number of measurements. The computational resources for the postprocessing required to reconstruct the state with high fidelity from these measurements is polynomial in the system size.
Quantum state tomography (QST) for reconstructing pure states requires exponentially increasing resources and measurements with the number of qubits by using state-of-the-art quantum compressive sensing (CS) methods. In this article, QST reconstruction for any pure state composed of the superposition of K different computational basis states of n qubits in a specific measurement set-up, i.e., denoted as K-sparse, is achieved without any initial knowledge and with quantum polynomial-time complexity of resources based on the assumption of the existence of polynomial size quantum circuits for implementing exponentially large powers of a specially designed unitary operator. The algorithm includes O(2/∣ck∣2) repetitions of conventional phase estimation algorithm depending on the probability ∣ck∣2 of the least possible basis state in the superposition and O(dK(logK)c) measurement settings with conventional quantum CS algorithms independent from the number of qubits while dependent on K for constant c and d. Quantum phase estimation algorithm is exploited based on the favorable eigenstructure of the designed operator to represent any pure state as a superposition of eigenvectors. Linear optical set-up is presented for realizing the special unitary operator which includes beam splitters and phase shifters where propagation paths of single photon are tracked with which-path-detectors. Quantum circuit implementation is provided by using only CNOT, phase shifter and −π/2 rotation gates around X-axis in Bloch sphere, i.e., RX(−π/2), allowing to be realized in NISQ devices. Open problems are discussed regarding the existence of the unitary operator and its practical circuit implementation.
Quantum state tomography (QST) is a central technique to fully characterize an unknown quantum state. However, standard QST requires an exponentially growing number of quantum measurements against the system size, which limits its application to smaller systems. Here, we explore the sparsity of underlying quantum state and propose a QST scheme that combines the matrix product states’ representation of the quantum state with a supervised machine learning algorithm. Our method could reconstruct the unknown sparse quantum states with very high precision using only a portion of the measurement data in a randomly selected basis set. In particular, we demonstrate that the Wolfgang states could be faithfully reconstructed using around 25% of the whole basis, and that the randomly generated quantum states, which could be efficiently represented as matrix product states, could be faithfully reconstructed using a number of bases that scales sub-exponentially against the system size.