Kód: 54579829
Reactive PublishingAccelerate atomic-scale simulations and navigate vast chemical spaces with machine learning.Traditional density functional theory (DFT) calculations offer high quantum-mechanical accuracy for predicting material ... celý popis
Angličtina
Zadejte do formuláře e-mailovou adresu a jakmile knihu naskladníme, zašleme vám o tom zprávu. Pohlídáme vše za vás.
Nákupem získáte 100 bodů
Anotace knihy
Reactive Publishing
Accelerate atomic-scale simulations and navigate vast chemical spaces with machine learning.
Traditional density functional theory (DFT) calculations offer high quantum-mechanical accuracy for predicting material properties, but their computational cost restricts simulations to small systems and short timescales. Machine Learning Potentials for Materials Discovery with Python provides a practical, end-to-end engineering framework for bridging the gap between ab initio accuracy and large-scale molecular dynamics.
Written for computational materials scientists, physical chemists, and Python developers working in scientific computing, this guide walks through the design, validation, and deployment of interatomic potentials using modern open-source ecosystems.
Inside, you will learn how to:
Construct Surrogate Models: Transform atomic environments into rotationally and permutationally invariant descriptors, and train neural network and kernel-based potentials on DFT datasets.
Implement Active Learning Workflows: Automate the detection of high-uncertainty atomic configurations during simulations, iteratively querying reference calculations to expand training sets without manual curation.
Scale High-Throughput Screening: Integrate trained potentials into molecular dynamics and structure-relaxation pipelines to rapidly filter candidate crystals, alloys, and catalysts.
Validate Physical Fidelity: Benchmark energy and force predictions, verify phonon dispersions and elastic constants, and prevent unphysical extrapolation in out-of-distribution structures.
Grounded in reproducible Python workflows, this book equips researchers and practitioners with the tools required to replace brute-force quantum simulations with fast, reliable machine learning pipelines.
Parametry knihy
1004 Kč
AngličtinaOsobní odběr Praha, Brno a 48091 dalších
Copyright ©2008-26 nejlevnejsi-knihy.cz Všechna práva vyhrazenaSoukromíCookies
Vrácení do měsíce
571 999 099 (8-15.30h)Nákupní košík ( prázdný )