Machine Learning Interatomic Potentials and Long-Range Physics

The Journal of Physical Chemistry A, Journal Year: 2023, Volume and Issue: 127(11), P. 2417 - 2431

Published: Feb. 21, 2023

Advances in machine learned interatomic potentials (MLIPs), such as those using neural networks, have resulted short-range models that can infer interaction energies with near ab initio accuracy and orders of magnitude reduced computational cost. For many atom systems, including macromolecules, biomolecules, condensed matter, model become reliant on the description short- long-range physical interactions. The latter terms be difficult to incorporate into an MLIP framework. Recent research has produced numerous considerations for nonlocal electrostatic dispersion interactions, leading a large range applications addressed MLIPs. In light this, we present Perspective focused key methodologies being used where presence physics chemistry are crucial describing system properties. strategies covered include MLIPs augmented corrections, electrostatics calculated charges predicted from atomic environment descriptors, use self-consistency message passing iterations propagated information, obtained via equilibration schemes. We aim provide pointed discussion support development learning-based systems contributions only nearsighted deficient.

Language: Английский

---

Nested sampling for physical scientists

Nature Reviews Methods Primers, Journal Year: 2022, Volume and Issue: 2(1)

Published: May 26, 2022

Language: Английский

---

Design of a new optimized U-shaped lightweight liquid-cooled battery thermal management system for electric vehicles: A machine learning approach

International Communications in Heat and Mass Transfer, Journal Year: 2022, Volume and Issue: 136, P. 106209 - 106209

Published: June 25, 2022

Language: Английский

---

3D‐Printed Functional Hydrogel by DNA‐Induced Biomineralization for Accelerated Diabetic Wound Healing

Advanced Science, Journal Year: 2023, Volume and Issue: 10(17)

Published: April 19, 2023

Chronic wounds in diabetic patients are challenging because their prolonged inflammation makes healing difficult, thus burdening patients, society, and health care systems. Customized dressing materials needed to effectively treat such that vary shape depth. The continuous development of 3D-printing technology along with artificial intelligence has increased the precision, versatility, compatibility various materials, providing considerable potential meet abovementioned needs. Herein, functional inks comprising DNA from salmon sperm DNA-induced biosilica inspired by marine sponges, developed for machine learning-based wound dressings. biomineralized silica incorporated into hydrogel a fast, facile manner. 3D-printed generates provided appropriate porosity, characterized effective exudate blood absorption at sites, mechanical tunability indicated good fidelity printability during optimized 3D printing. Moreover, act as nanotherapeutics, enhancing biological activity dressings terms reactive oxygen species scavenging, angiogenesis, anti-inflammation activity, thereby accelerating acute healing. These bioinspired hydrogels produce using biomineralization strategy an excellent platform clinical applications chronic repair.

Language: Английский

---

New Opportunity: Machine Learning for Polymer Materials Design and Discovery

Advanced Theory and Simulations, Journal Year: 2022, Volume and Issue: 5(5)

Published: Feb. 12, 2022

Abstract Under the guidance of material genome initiative (MGI), use data‐driven methods to discover new materials has become an innovation science. The polymer have been one most important parts in science for excellent physical and chemical properties as well corresponding complex structures. Machine learning, core methods, taken place design discovery. In this review, authors introduced applications machine learning discovery materials. development tendency published papers about materials, commonly used algorithms, descriptors, workflow recent progresses are summarized. Then, detail how assist is fully discussed combined with two cases. Finally, opportunities challenges on future prospects field proposed.

Language: Английский

---

Machine learning potentials for metal-organic frameworks using an incremental learning approach

npj Computational Materials, Journal Year: 2023, Volume and Issue: 9(1)

Published: Feb. 6, 2023

Abstract Computational modeling of physical processes in metal-organic frameworks (MOFs) is highly challenging due to the presence spatial heterogeneities and complex operating conditions which affect their behavior. Density functional theory (DFT) may describe interatomic interactions at quantum mechanical level, but computationally too expensive for systems beyond nanometer picosecond range. Herein, we propose an incremental learning scheme construct accurate data-efficient machine potentials MOFs. The builds on power equivariant neural network combination with parallelized enhanced sampling on-the-fly training simultaneously explore learn phase space iterative manner. With only a few hundred single-point DFT evaluations per material, transferable are obtained, even flexible multiple structurally different phases. universally applicable pave way model framework materials larger spatiotemporal windows higher accuracy.

Language: Английский

---

Machine Learning: A New Paradigm in Computational Electrocatalysis

The Journal of Physical Chemistry Letters, Journal Year: 2022, Volume and Issue: 13(34), P. 7920 - 7930

Published: Aug. 18, 2022

Designing and screening novel electrocatalysts, understanding electrocatalytic mechanisms at an atomic level, uncovering scientific insights lie the center of development electrocatalysis. Despite certain success in experiments computations, it is still difficult to achieve above objectives due complexity systems vastness chemical space for candidate electrocatalysts. With advantage machine learning (ML) increasing interest electrocatalysis energy conversion storage, data-driven research motivated by artificial intelligence (AI) has provided new opportunities discover promising investigate dynamic reaction processes, extract knowledge from huge data. In this Perspective, we summarize recent applications ML electrocatalysis, including electrocatalysts simulation processes. Furthermore, interpretable methods are discussed accelerate generation. Finally, blueprint envisaged future

Language: Английский

---

How to validate machine-learned interatomic potentials

The Journal of Chemical Physics, Journal Year: 2023, Volume and Issue: 158(12)

Published: March 2, 2023

Machine learning (ML) approaches enable large-scale atomistic simulations with near-quantum-mechanical accuracy. With the growing availability of these methods, there arises a need for careful validation, particularly physically agnostic models-that is, potentials that extract nature atomic interactions from reference data. Here, we review basic principles behind ML and their validation atomic-scale material modeling. We discuss best practice in defining error metrics based on numerical performance, as well guided validation. give specific recommendations hope will be useful wider community, including those researchers who intend to use materials "off shelf."

Language: Английский

---

Evaluation of the MACE force field architecture: From medicinal chemistry to materials science

The Journal of Chemical Physics, Journal Year: 2023, Volume and Issue: 159(4)

Published: July 28, 2023

The MACE architecture represents the state of art in field machine learning force fields for a variety in-domain, extrapolation, and low-data regime tasks. In this paper, we further evaluate by fitting models published benchmark datasets. We show that generally outperforms alternatives wide range systems, from amorphous carbon, universal materials modeling, general small molecule organic chemistry to large molecules liquid water. demonstrate capabilities model on tasks ranging constrained geometry optimization molecular dynamics simulations find excellent performance across all tested domains. is very data efficient can reproduce experimental vibrational spectra when trained as few 50 randomly selected reference configurations. strictly local atom-centered sufficient such even case weakly interacting assemblies.

Language: Английский

---

Hyperactive learning for data-driven interatomic potentials

npj Computational Materials, Journal Year: 2023, Volume and Issue: 9(1)

Published: Sept. 13, 2023

Abstract Data-driven interatomic potentials have emerged as a powerful tool for approximating ab initio potential energy surfaces. The most time-consuming step in creating these is typically the generation of suitable training database. To aid this process hyperactive learning (HAL), an accelerated active scheme, presented method rapid automated database assembly. HAL adds biasing term to physically motivated sampler (e.g. molecular dynamics) driving atomic structures towards uncertainty turn generating unseen or valuable configurations. proposed framework used develop cluster expansion (ACE) AlSi10 alloy and polyethylene glycol (PEG) polymer starting from roughly dozen initial generated ACE are shown be able determine macroscopic properties, such melting temperature density, with close experimental accuracy.

Language: Английский

---