问题

Please output this exact text, with no changes at all except for fixing the misspellings. Please leave all other stylistic decisions like commas and US vs British spellings as in the original text.

The interaction of condensed phase systems whith exerternalelecticfiels is crutial in myriad processes in nature adn technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and convertion systems includng supercapacitors, batteries adn solar cells. Molecular simulation in the presense of electric fields owudl give importamt atomistic insight into theese processes but applicaitons ofther most accurate methods such as ab-initio molecular dynamics are limited in scope byther computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push bcakhten accessable time adn length scales of such simulations at virtualy no loss in acuracy. The total forces onhten atoms are expressed in terms ofhten unperturbed potential energy surfce represented by a standard neural network potential anbd afiel-induced pertubation obtained frome a series expansion ofhten field interaction truncated atfirts order. The latter is represented in terms of an equivariant graph neural network, trained onther atomic polar tensor. PNNP MD is shown to give excellent results forhten dielectric relaxation dynamics, the dielectric constant andhtenfiel-dependent IR spectum of liquid water whn compared to ab-initio molecular dynamics or experiment, up to suprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approachs, tje twpo neural networks on hwihc PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating tyhathten networksonotlonly interpolate but alsot reliably extrapolate thefiel response. PNNP MD is based on rigorous theory yet it is simple, general, modular, anbd systematically improvable alowing us to obtain atomistic insight intohten interaction of a wide range of condensed phase systems with exerternalelectic fields.

答案

模型 模型答案 正确答案
金声玉亮2.0 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology, ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries, and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes, but applications of the most accurate methods, such as ab-initio molecular dynamics, are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant, and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations, demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable, allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
deepseek-r1

The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/Å. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.		 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
gpto1 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/Å. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
gpt-4o-2024-08-06 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
DeepSeek-Coder-V2-Lite-Instruct The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes, but applications of the most accurate methods such as ab-initio molecular dynamics are limited by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations with virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant, and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
Meta-Llama-3.1-405B-Instruct-Turbo Here is the text with the misspellings corrected:

The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.		 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
Meta-Llama-3.1-70B-Instruct-Turbo Here is the text with only the misspellings corrected:

The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.		 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
Qwen2-72B-Instruct The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology, ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries, and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes, but applications of the most accurate methods, such as ab-initio molecular dynamics, are limited in scope by the computational expense. Here, we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant, and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations, demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable, allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
baichuan4 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
doubao The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
gpto1 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/Å. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
kimi The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries, and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant, and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
minimax The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries, and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes, but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant, and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/Å. This is remarkable because, The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
wenxinyiyan The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
zhipu The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by the computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.
deepseek-v3 The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of their most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields. The interaction of condensed phase systems with external electric fields is crucial in myriad processes in nature and technology ranging from the field-directed motion of cells (galvanotaxis), to energy storage and conversion systems including supercapacitors, batteries and solar cells. Molecular simulation in the presence of electric fields would give important atomistic insight into these processes but applications of the most accurate methods such as ab-initio molecular dynamics are limited in scope by their computational expense. Here we introduce Perturbed Neural Network Potential Molecular Dynamics (PNNP MD) to push back the accessible time and length scales of such simulations at virtually no loss in accuracy. The total forces on the atoms are expressed in terms of the unperturbed potential energy surface represented by a standard neural network potential and a field-induced perturbation obtained from a series expansion of the field interaction truncated at first order. The latter is represented in terms of an equivariant graph neural network, trained on the atomic polar tensor. PNNP MD is shown to give excellent results for the dielectric relaxation dynamics, the dielectric constant and the field-dependent IR spectrum of liquid water when compared to ab-initio molecular dynamics or experiment, up to surprisingly high field strengths of about 0.2 V/A. This is remarkable because, in contrast to most previous approaches, the two neural networks on which PNNP MD is based are exclusively trained on zero-field molecular configurations demonstrating that the networks not only interpolate but also reliably extrapolate the field response. PNNP MD is based on rigorous theory yet it is simple, general, modular, and systematically improvable allowing us to obtain atomistic insight into the interaction of a wide range of condensed phase systems with external electric fields.