quantum mechanics calculations with learned estimators. This raises questions about the representability of complex quantum chemical systems with neural networks. Can local-variable models efficiently approximate nonlocal quantum chemical features? Here, we find that convolutional architectures, those that only aggregate information locally, cannot efficiently represent aromaticity and conjugation in large systems. They cannot represent long-range nonlocality known to be important in quantum chemistry. This study uses aromatic and conjugated systems computed from molecule graphs, though reproducing quantum simulations is the ultimate goal. This task, by definition, is both computable and known to be important to chemistry. The failure of convolutional architectures on this focused task calls into question their use in modeling quantum mechanics. To remedy this heretofore unrecognized deficiency, we introduce a new architecture that propagates information back and forth in waves of nonlinear computation. This architecture is still a local-variable model, and it is both computationally and representationally efficient, processing molecules in sublinear time with far fewer parameters than convolutional networks. Wave-like propagation models aromatic and conjugated systems with high accuracy, and even models the impact of small structural changes on large molecules. This new architecture demonstrates that some nonlocal features of quantum chemistry can be efficiently represented in local variable models.
quantum mechanics calculations with learned estimators. This raises questions about the representability of complex quantum chemical systems with neural networks. Can local-variable models efficiently approximate nonlocal quantum chemical features? Here, we find that convolutional architectures, those that only aggregate information locally, cannot efficiently represent aromaticity and conjugation in large systems. They cannot represent long-range nonlocality known to be important in quantum chemistry. This study uses aromatic and conjugated systems computed from molecule graphs, though reproducing quantum simulations is the ultimate goal. This task, by definition, is both computable and known to be important to chemistry. The failure of convolutional architectures on this focused task calls into question their use in modeling quantum mechanics. To remedy this heretofore unrecognized deficiency, we introduce a new architecture that propagates information back and forth in waves of nonlinear computation. This architecture is still a local-variable model, and it is both computationally and representationally efficient, processing molecules in sublinear time with far fewer parameters than convolutional networks. Wave-like propagation models aromatic and conjugated systems with high accuracy, and even models the impact of small structural changes on large molecules. This new architecture demonstrates that some nonlocal features of quantum chemistry can be efficiently represented in local variable models.
We present a graph-based machine-learning framework for simulating the time evolution of electronic wavefunctions and densities in quantum systems. Inspired by parallels between time-dependent quantum propagators and spectral graph convolutions, we employ a recursive Chebyshev graph neural network architecture capable of learning the dynamics of electronic processes across a range of external potentials and Hamiltonians. Two model variants are introduced: one that evolves the full complex-valued wavefunction and the other that propagates only the electron density. Both models are trained on trajectory data generated from tight-binding Hamiltonians and a time-dependent electron-phonon coupled system. Our results demonstrate that wavefunction-based models achieve near-exact long-time propagation across static and dynamic regimes, while density-only models maintain strong performance using physics-informed loss functions, even in the absence of phase information. This work lays the foundation for coarse-grained, resolution-independent propagators for electronic dynamics and opens new pathways for scalable quantum simulations in complex molecular and condensed-phase systems.
Machine learning has made significant contributions to the fields of chemistry and materials science, enabling the exploration of vast chemical space through large-scale quantum chemical calculations. These models provide fast and accurate predictions of atomistic chemical properties, but they have limitations when it comes to capturing the electronic degrees of freedom of a molecule. This restricts their application in reactive chemistry and chemical analysis. To address this limitation, we introduce a deep learning framework that predicts the quantum mechanical wavefunction of a molecule in a local basis of atomic orbitals. The wavefunction serves as a foundational representation from which all other ground-state properties can be derived. Our approach maintains complete access to the electronic structure through the wavefunction, while achieving computational efficiency comparable to force-field methods. Moreover, the framework captures quantum mechanics in a form that can be analytically differentiated, allowing for efficient optimization and exploration of chemical systems. We demonstrate the potential of our approach through several examples. By leveraging the predicted wavefunction, we showcase the ability to perform inverse design of molecular structures to target specific electronic property optimizations. This opens exciting avenues for tailoring molecular structures to achieve desired electronic characteristics. Additionally, our framework paves the way for enhanced synergy between machine learning and quantum chemistry, enabling more comprehensive investigations into complex chemical systems.
Neural scaling laws are driving the machine learning community toward training ever-larger foundation models across domains, assuring high accuracy and transferable representations for extrapolative tasks. We test this promise in quantum chemistry by scaling model capacity and training data from quantum chemical calculations. As a generalization task, we evaluate the resulting models' predictions of the bond dissociation energy of neutral H2, the simplest possible molecule. We find that, regardless of dataset size or model capacity, models trained only on stable structures fail dramatically to even qualitatively reproduce the H2 energy curve. Only when compressed and stretched geometries are explicitly included in training do the predictions roughly resemble the correct shape. Nonetheless, the largest foundation models trained on the largest and most diverse datasets containing dissociating diatomics exhibit serious failures on simple diatomic molecules. Most strikingly, they cannot reproduce the trivial repulsive energy curve of two bare protons, revealing their failure to learn the basic Coulomb's law involved in electronic structure theory. These results suggest that scaling alone is insufficient for building reliable quantum chemical models.
Neural network quantum states represent a powerful approach for solving electronic structures in strongly correlated molecular and material systems. For a neural network ansatz to be accurate, it must effectively learn the phase of a complex wave function. In this work, we demonstrate several different network structures as the phase network for a Transformer-based neural network quantum state implementation. We compare the accuracy of ground state energies, the number of parameters, and computational time across several small molecules. Furthermore, we propose a phase network setup that combines cross attention and multilayer perceptron structures, with the number of parameters remaining constant across different systems. Such an architecture may help reduce computational costs and enable transfer learning to larger quantum chemical systems.
The field of quantum chemistry has long been defined by the central computational challenge of the quantum many-body fermionic system problem, main the electronic Schrödinger equation. This exponential scaling makes it computationally intractable to solve, the exact governing electronic Schrödinger equation. Approximation methods have been developed, ranging from the widely used Density Functional Theory (DFT) to Coupled Cluster (CC) theory. While powerful, these conventional approaches often involve controlled truncations which caused a trade-off between accuracy, leaving many systems—particularly those with strong electron correlation—beyond the reach of ab initio simulation. Furthermore, the famous fermionic sign oscillation problem is particularly severe in stochastic methods like Quantum Monte Carlo (QMC). 1 Traditional Ansätze are also frequently engineered specifically for particular systems, which results in a fragmentation of results and fundamentally limits the transferability. Recent research demonstrates that with carefully designed architectures, ANNs can not only represent ground states of model systems with high precision but also tackle the sign problem, offering a promising path forward for strongly correlated electronic structure calculations.2,3 This review charts the paradigm shift driven by Neural Network Quantum States (NNQS), which leverage the representational power of deep learning to overcome these barriers. We detail the architectural evolution from Restricted Boltzmann Machines to autoregressive models like RNNs and Transformers, which enable exact, uncorrelated sampling and bypass critical bottlenecks.2,4,5Furthermore, we explore the geometric foundation of high-efficiency optimization via the neural networks, which decouples optimization cost from model complexity.6 The frontier of the field is marked by the integration of operator learning and hybrid quantum-classical frameworks, such as Reinforcement Learning for contracted quantum eigen-solvers and Quantum-Enhanced Neural Networks, which leverage quantum processors to enhance expressivity for hardware-efficient Ansätze.7,8 By framing the problem through the lens of computational resource allocation—specifically, trading exponential memory demands for polynomialcomplexity optimization and sampling—this review elucidates how recent advances at the intersection of artificial neural networks and quantum computing have evolved. The emergence of those methods not merely progress into a powerful simulation-tools, but also a hybrid computational interface.9 This convergence offers a versatile and systematically improvable framework for potentially addressing some of the most challenging problems in quantum physics and chemistry.
The field of quantum chemistry has long been defined by the central computational challenge of the quantum many-body fermionic system problem, main the electronic Schrödinger equation. This exponential scaling makes it computationally intractable to solve, the exact governing electronic Schrödinger equation. Approximation methods have been developed, ranging from the widely used Density Functional Theory (DFT) to Coupled Cluster (CC) theory. While powerful, these conventional approaches often involve controlled truncations which caused a trade-off between accuracy, leaving many systems—particularly those with strong electron correlation—beyond the reach of ab initio simulation. Furthermore, the famous fermionic sign oscillation problem is particularly severe in stochastic methods like Quantum Monte Carlo (QMC). 1 Traditional Ansätze are also frequently engineered specifically for particular systems, which results in a fragmentation of results and fundamentally limits the transferability. Recent research demonstrates that with carefully designed architectures, ANNs can not only represent ground states of model systems with high precision but also tackle the sign problem, offering a promising path forward for strongly correlated electronic structure calculations.2,3 This review charts the paradigm shift driven by Neural Network Quantum States (NNQS), which leverage the representational power of deep learning to overcome these barriers. We detail the architectural evolution from Restricted Boltzmann Machines to autoregressive models like RNNs and Transformers, which enable exact, uncorrelated sampling and bypass critical bottlenecks.2,4,5Furthermore, we explore the geometric foundation of high-efficiency optimization via the neural networks, which decouples optimization cost from model complexity.6 The frontier of the field is marked by the integration of operator learning and hybrid quantum-classical frameworks, such as Reinforcement Learning for contracted quantum eigen-solvers and Quantum-Enhanced Neural Networks, which leverage quantum processors to enhance expressivity for hardware-efficient Ansätze.7,8 By framing the problem through the lens of computational resource allocation—specifically, trading exponential memory demands for polynomialcomplexity optimization and sampling—this review elucidates how recent advances at the intersection of artificial neural networks and quantum computing have evolved. The emergence of those methods not merely progress into a powerful simulation-tools, but also a hybrid computational interface.9 This convergence offers a versatile and systematically improvable framework for potentially addressing some of the most challenging problems in quantum physics and chemistry.