Abstract The lottery ticket hypothesis (LTH) posits that within overparametrized neural networks, there exist sparse subnetworks that are capable of matching the performance of the original model when trained in isolation from the original initialization. We extend this hypothesis to the unsupervised task of approximating the ground state of quantum many-body Hamiltonians, a problem equivalent to finding a neural-network compression of the lowest-lying eigenvector of an exponentially large matrix. Focusing on two representative quantum Hamiltonians, the transverse field Ising model and the toric code, we demonstrate that sparse neural networks can reach accuracies comparable to their dense counterparts, even when pruned by more than an order of magnitude in parameter count. Crucially, and unlike the original LTH, we find that performance depends only on the structure of the sparse subnetwork, not on the specific initialization, when trained in isolation. Moreover, we identify universal scaling behavior that persists across network sizes and physical models, where the boundaries of scaling regions are determined by the underlying Hamiltonian. At the onset of high-error scaling, we observe signatures of a sparsity-induced quantum phase transition that is first-order in shallow networks. Finally, we demonstrate that pruning enhances interpretability by linking the structure of sparse subnetworks to the underlying physics of the Hamiltonian.
Abstract The lottery ticket hypothesis (LTH) posits that within overparametrized neural networks, there exist sparse subnetworks that are capable of matching the performance of the original model when trained in isolation from the original initialization. We extend this hypothesis to the unsupervised task of approximating the ground state of quantum many-body Hamiltonians, a problem equivalent to finding a neural-network compression of the lowest-lying eigenvector of an exponentially large matrix. Focusing on two representative quantum Hamiltonians, the transverse field Ising model and the toric code, we demonstrate that sparse neural networks can reach accuracies comparable to their dense counterparts, even when pruned by more than an order of magnitude in parameter count. Crucially, and unlike the original LTH, we find that performance depends only on the structure of the sparse subnetwork, not on the specific initialization, when trained in isolation. Moreover, we identify universal scaling behavior that persists across network sizes and physical models, where the boundaries of scaling regions are determined by the underlying Hamiltonian. At the onset of high-error scaling, we observe signatures of a sparsity-induced quantum phase transition that is first-order in shallow networks. Finally, we demonstrate that pruning enhances interpretability by linking the structure of sparse subnetworks to the underlying physics of the Hamiltonian.
Quantum computing is an emerging field in computer science that has seen considerable progress in recent years, especially in machine learning. By harnessing the principles of quantum physics, it can surpass the limitations of classical algorithms. However, variational quantum circuits (VQCs), which rely on adjustable parameters, often face the barren plateau phenomenon, hindering optimization. The Lottery Ticket Hypothesis (LTH) is a recent concept in classical machine learning that has led to notable improvements in parameter efficiency for neural networks. It states that within a large network, a smaller, more efficient subnetwork, or ''winning ticket,'' can achieve comparable performance, potentially circumventing plateau challenges. In this work, we investigate whether this idea can apply to VQCs. We show that the weak LTH holds for VQCs, revealing winning tickets that retain just 26.0\% of the original parameters. For the strong LTH, where a pruning mask is learned without any training, we discovered a winning ticket in a binary VQC, achieving 100\% accuracy with only 45\% of the weights. These findings indicate that LTH may mitigate barren plateaus by reducing parameter counts while preserving performance, thus enhancing the efficiency of VQCs in quantum machine learning tasks.
Quantum machine learning (QML) aims to accelerate machine learning tasks by exploiting quantum computation. Previous work studied a QML algorithm for selecting sparse subnetworks from large shallow neural networks. Instead of directly solving an optimization problem over a large-scale network, this algorithm constructs a sparse subnetwork by sampling hidden nodes from an optimized probability distribution defined using the ridgelet transform. The quantum algorithm performs this sampling in time O(D) in the data dimension D, whereas a naive classical implementation relies on handling exponentially many candidate nodes and hence takes exp[O(D)] time. In this work, we construct and analyze a quantum-inspired fully classical algorithm for the same sampling task. We show that our algorithm runs in time O(poly(D)), thereby removing the exponential dependence on D from the previous classical approach. Numerical simulations show that the proposed sampler achieves empirical risk comparable to exact sampling from the optimized distribution and substantially lower than sampling from the non-optimized uniform distribution, while also exhibiting exponentially improved runtime scaling compared with the conventional classical implementation. These successful dequantization results show that sparse subnetwork selection via optimized sampling can be achieved classically with polynomial data-dimension scaling on conventional computers without quantum hardware, providing an alternative to the existing quantum algorithm.
Quantum Neural Networks (QNNs) possess high expressivity via rich unitary transformations, offering abundant representational capacity while maintaining a parameter-effective architecture. In classical deep learning, the high expressivity of deep neural networks (DNNs) motivates the Strong Lottery Ticket Hypothesis (SLTH), which suggests that a randomly initialized DNN contains sparse subnetworks that achieve high performance immediately or serve as a warm start. On the quantum side, a key question is: Does a strong lottery ticket exist in QNNs? Our answer is, Quantum-Strong Lottery Ticket Hypothesis (Q-SLTH), i.e., the first theoretical analysis for the existence of high-performing subnetworks in random scratch of QNNs. We validate our theory by proposing a quantum edge-popup algorithm to find strong subnetworks and empirically identifying winning tickets in randomly initialized QNNs on the MNIST and Fashion-MNIST benchmarks. For practice, Q-SLTH suggests a way to compress QNNs, allowing them to use fewer gates and thereby mitigating quantum-noise-induced degradation that is amplified across multiple layers of quantum gates.
Neural quantum states (NQS) have emerged as powerful representations of quantum states with rapidly expanding applications across quantum many-body physics. Yet our understanding of when neural networks can efficiently represent physical quantum states remains limited, in part due to the nonlinear parameterization of NQS, the intricate sign structure of quantum states, and the sensitive dependence on basis choice. We introduce Walshness, a complexity metric of a quantum state which quantifies whether it admits a compact neural-network representation. We rigorously prove that a general quantum many-body state admits an efficient NQS representation if it has bounded Walshness, and vice versa, for various NQS architectures. We provide an algorithm for finding the optimal basis by minimizing the energy of a corresponding classical spin model. This allows us to empirically improve the learnability of various physical quantum states, including ground states of the transverse-field Ising model and mixed-field toric code, often reducing infidelity by orders of magnitude, and generalizes the basis choice given by the well-known Marshall sign rule in frustrated antiferromagnets. Our results provide an intrinsic complexity metric for quantum states with provable connections to their neural-network representability.
Variational Quantum Algorithms (VQAs) face two compounding obstacles on Noisy Intermediate-Scale Quantum (NISQ) hardware: the barren plateau phenomenon, wherein gradient variancevanishes exponentially with system size n, and device decoherence, which degrades coherent signalbefore classical optimizers can act. We introduce the Sparse Modulated Quantum Neural Network(SMQNN), a hardware-native ansatz that resolves both obstacles through a single structural principle:replacing global entanglement with block-diagonal modules of fixed width k, interconnected exclusivelyby classical Mid-Circuit Measurements (MCMs). We prove that the gradient variance of SMQNN islower-bounded by Ω(∥Wm∥22· 4−k), independent of total system size n, thereby providing a provableescape from exponential trainability collapse. We further prove that classical state preparation atmodule boundaries resets depolarization entropy, reducing the effective error-accumulation lengthfrom k to keff = k(1 − fmcm/2). This yields an exponential improvement in the signal-to-noiseratio parameterized by the reset fraction fmcm, with similar principles applying to trapped-ion andneutral-atom error models as validated by simulation. We validate SMQNN empirically on a physical156-qubit IBM Marrakesh Heron-architecture processor using 4096 shots per experiment and TwirledReadout Error Extinction. Across three physically distinct benchmarks—topological phase transitionsin the Su-Schrieffer-Heeger (SSH) model, Many-Body Localization (MBL) imbalance dynamics,and Quantum Error Correction (QEC) syndrome detection—SMQNN consistently outperformsdepth-matched dense-VQC baselines. Edge parity accuracy improved from 0.83 (dense) to 0.90(SMQNN); syndrome correction fidelity improved from 45% to 80%. We explicitly characterizethe expressivity trade-off: SMQNN achieves a 28× reduction in ground-state accuracy for globallyentangled Hamiltonians, confirming its intended domain as topological, localized, or modular quantumphenomena. We systematically characterize the expressivity-noise trade-off, demonstrating thatSMQNN intentionally restricts Hilbert space to provide robust scaling specifically for spatiallylocal problems. Finally, we demonstrate platform-agnostic generality via simulations of trapped-ionand neutral-atom noise regimes, expanding the SMQNN architecture into a comprehensive toolkitfor NISQ algorithm design. All circuits, gate counts, and IBM Job IDs are open-sourced for fullreproducibility.
In recent years, generative artificial neural networks based on restricted Boltzmann machines (RBMs) have been successfully employed as accurate and flexible variational wave functions for clean quantum many-body systems. In this article, we explore their use in simulations of disordered quantum Ising chains. The standard dense RBM with all-to-all interlayer connectivity is not particularly appropriate for large disordered systems, since in such systems one cannot exploit translational invariance to reduce the amount of parameters to be optimized. To circumvent this problem, we implement sparse RBMs, whereby the visible spins are connected only to a subset of local hidden neurons, thus reducing the amount of parameters. We assess the performance of sparse RBMs as a function of the range of the allowed connections, and we compare it with that of dense RBMs. Benchmark results are provided for two sign-problem-free Hamiltonians, namely pure and random quantum Ising chains. The RBM Ansätzes are trained using the unsupervised learning scheme based on projective quantum Monte Carlo (PQMC) algorithms. We find that the sparse connectivity facilitates the training process and allows sparse RBMs to outperform their dense counterparts. Furthermore, the use of sparse RBMs as guiding functions for PQMC simulations allows us to perform PQMC simulations at a reduced computational cost, avoiding possible biases due to finite random-walker populations. We obtain unbiased predictions for the ground-state energies and the magnetization profiles with fixed boundary conditions, at the ferromagnetic quantum critical point. The magnetization profiles agree with the Fisher-de Gennes scaling relation for conformally invariant systems, including the scaling dimension predicted by the renormalization-group analysis.