Search

In this report, we propose a novel quantum diagonalization algorithm through optimizing a variational quantum circuit. The optimization process utilizes measurement outcomes in the computational basis to construct an objective function for variational optimization. Two distinct objective functions are introduced in this report. The first is constructed from the probabilities of 2^N computational basis states. By maximizing the sum of the squared diagonal elements of the evolved N-qubit quantum state, this function drives the state toward a diagonal form, with the optimized value corresponding to the purity of the input quantum state. To address the experimental challenges associated with measuring all 2^N basis states, we propose a second objective function based on N single-qubit measurement probabilities. This reduces measurement complexity with exponential growth, enhancing the feasibility of experimental implementation. In both cases, we explore the relationship between measurement probability distributions and the diagonalization of the evolved quantum state. Numerical simulations and analytical insights demonstrate that the variational optimization effectively transforms the input quantum state into its diagonalized form, offering a practical framework for quantum state diagonalization., Comment: 7 pages, 7 figures

The rapid development of quantum technology demands efficient characterization of complex quantum many-body states. However, full quantum state tomography requires an exponential number of measurements in system size, preventing its practical use in large-scale quantum devices. A major recent breakthrough in this direction, called classical shadow tomography, significantly reduces the sample complexity, the number of samples needed to estimate properties of a state, by implementing random Clifford rotations before measurements. Despite many recent efforts, reducing the sample complexity below $\mathbf{2^k}$ for extracting any non-successive local operators with a size $\sim \mathbf{k}$ remains a challenge. In this work, we achieve a significantly smaller sample complexity of $\mathbf{\sim 1.8^k}$ using a protocol that hybridizes locally random and globally deterministic unitary operations. The key insight is the discovery of a deterministic global unitary, termed as \textit{contractive unitary}, which is more efficient in reducing the operator size to enhance tomography efficiency. The contractive unitary perfectly matches the advantages of the atom array quantum computation platform and is readily realized in the atom array quantum processor. More importantly, it highlights a new strategy in classical shadow tomography, demonstrating that a random-deterministic hybridized protocol can be more efficient than fully random measurements.

In this work, we report on a novel quantum state reconstruction process based on the disentanglement algorithm. Using variational quantum circuits, we disentangle the quantum state to a product of computational zero states. Inverse evolution of the zero states reconstructs the quantum state up to an overall phase. By sequentially disentangling the qubit one by one, we reduce the required measurements with only single qubit measurement. Demonstrations with our proposal for the reconstruction of the random states are presented where variational quantum circuit is optimized by disentangling process. To facilitate experimental implementation, we also employ reinforcement learning for quantum circuit design with limited discrete quantum gates. Our method is universal and imposes no specific ansatz or constrain on the quantum state., Comment: 7 pages, 9 figures

As a hybrid of artificial intelligence and quantum computing, quantum neural networks (QNNs) have gained significant attention as a promising application on near-term, noisy intermediate-scale quantum (NISQ) devices. Conventional QNNs are described by parametrized quantum circuits, which perform unitary operations and measurements on quantum states. In this work, we propose a novel approach to enhance the expressivity of QNNs by incorporating randomness into quantum circuits. Specifically, we introduce a random layer, which contains single-qubit gates sampled from an trainable ensemble pooling. The prediction of QNN is then represented by an ensemble average over a classical function of measurement outcomes. We prove that our approach can accurately approximate arbitrary target operators using Uhlmann's theorem for majorization, which enables observable learning. Our proposal is demonstrated with extensive numerical experiments, including observable learning, R\'enyi entropy measurement, and image recognition. We find the expressivity of QNNs is enhanced by introducing randomness for multiple learning tasks, which could have broad application in quantum machine learning., Comment: 6 pages, 4 figures

Preparation of quantum states is of vital importance for performing quantum computations and quantum simulations. In this work, we propose a general framework for preparing ground states of many-body systems by combining the measurement-feedback control process (MFCP) and the machine learning method. Using the Bayesian optimization (BO) strategy, the efficiency of determining the measurement and feedback operators in the MFCP is demonstrated. Taking the one dimensional Bose-Hubbard model as an example, we show that BO can generate optimal parameters, although constrained by the operator basis, which can drive the system to the low energy state with high probability in typical quantum trajectories.

Analytic continuation maps imaginary-time Green's functions obtained by various theoretical/numerical methods to real-time response functions that can be directly compared with experiments. Analytic continuation is an important bridge between many-body theories and experiments but is also a challenging problem because such mappings are ill-conditioned. In this work, we develop a neural network-based method for this problem. The training data is generated either using synthetic Gaussian-type spectral functions or from exactly solvable models where the analytic continuation can be obtained analytically. Then, we applied the trained neural network to the testing data, either with synthetic noise or intrinsic noise in Monte Carlo simulations. We conclude that the best performance is always achieved when a proper amount of noise is added to the training data. Moreover, our method can successfully capture multi-peak structure in the resulting response function for the cases with the best performance. The method can be combined with Monte Carlo simulations to compare with experiments on real-time dynamics.

In this work, we address the question whether a sufficiently deep quantum neural network can approximate a target function as accurate as possible. We start with simple but typical physical situations that the target functions are physical observables, and then we extend our discussion to situations that the learning targets are not directly physical observables, but can be expressed as physical observables in an enlarged Hilbert space with multiple replicas, such as the Loshimidt echo and the Renyi entropy. The main finding is that an accurate approximation is possible only when the input wave functions in the dataset do not exhaust the entire Hilbert space that the quantum circuit acts on, and more precisely, the Hilbert space dimension of the former has to be less than half of the Hilbert space dimension of the latter. In some cases, this requirement can be satisfied automatically because of the intrinsic properties of the dataset, for instance, when the input wave function has to be symmetric between different replicas. And if this requirement cannot be satisfied by the dataset, we show that the expressivity capabilities can be restored by adding one ancillary qubit where the wave function is always fixed at input. Our studies point toward establishing a quantum neural network analogy of the universal approximation theorem that lays the foundation for expressivity of classical neural networks.

In large-scale computation of physics problems, one often encounters the problem of determining a multi-dimensional function, which can be time-consuming when computing each point in this multi-dimensional space is already time-demanding. In the work, we propose that the active learning algorithm can speed up such calculations. The basic idea is to fit a multi-dimensional function by neural networks, and the key point is to make the query of labeled data economically by using a stratagem called "query by committee". We present the general protocol of this fitting scheme, as well as the procedure of how to further compute physical observables with the fitted functions. We show that this method can work well with two examples, which are quantum three-body problem in atomic physics and the anomalous Hall conductivity in condensed matter physics, respectively. In these examples, we show that one reaches an accuracy of few percent error for computing physical observables with less than $10\%$ of total data points compared with uniform sampling. With these two examples, we also visualize that by using the active learning algorithm, the required data are added mostly in the regime where the function varies most rapidly, which explains the mechanism for the efficiency of the algorithm. We expect broad applications of our method on various kind of computational physics problems., Comment: 7 pages, 8 figures

Recent experiment has observed Feshbach resonances between alkaline and alkaline-earth atoms. These Feshbach resonances are insensitive to the nuclear spin of alkaline-earth atoms. Ultilizing this feature, we propose to take this system as a candidate to perform quantum simulation of the Kondo effect. An alkaline atom can form a molecule with an alkaline-earth atom with different nuclear spins, which plays the role of spin-exchange scattering responsible for the Kondo effect. Furthermore, we point out that the existence of three-body bound state and atom-molecule resonance due to the Efimov effect can enhance this spin-exchange scattering, and therefore enhance the Kondo effect. We discuss this mechanism first with a three-body problem in free space, and then demonstrate that the same mechanism still holds when the alkaline atom is localized by an external trap and becomes an impurity embedded in the alkaline-earth atomic gases.

Motivated by recent experimental progresses, we investigate few-body properties of interacting spinless bosons nearby a d-wave resonance. Using the Skorniakov-Ter-Martirosion (STM) equations, we calculate the scattering length between an atom and a d-wave dimer, and we find that the atom-dimer scattering length is positive and is much smaller the result from the mean-field approximation. We also reveal unique properties of the three-body recombination rate for a degenerate Bose condensate nearby the d-wave resonance. We find that the total recombination rate is nearly a constant at the quasi-bound side, in contrast to the behavior of a thermal gas nearby high-partial wave resonance. We also find that the recombination rate monotonically increases across the unitary point toward the bound side, which is due to the largely enhanced coupling between the atom and the d-wave dimer with deeper binding energy. This monotonic behavior is also qualitatively different from that of a degenerate gas nearby an s-wave resonance counterpart.