相关实验视频
Updated: Jun 28, 2025
11:01
A Computer-assisted Multi-electrode Patch-clamp System
Published on: October 18, 2013
13.8K
积极学习和转移高维神经网络潜力转型金属的学习
Bilvin Varughese1,2, Sukriti Manna1,2, Troy D Loeffler1,2
1Department of Mechanical and Industrial Engineering, University of Illinois, Chicago, Illinois 60607, United States.
ACS applied materials & interfaces
|April 9, 2024
概括
机器学习潜力通过将转移和主动学习结合起来,准确地捕获纳米尺度和批量属性. 这种方法增强了材料设计的分子动力学模拟,加速了各种应用中的发现.
科学领域:
- 计算材料科学科学 计算材料科学
- 机器学习在化学和物理领域
- 凝聚物质物理学 凝聚物质物理学
背景情况:
- 经典分子动力学 (MD) 模拟对于材料建模至关重要,但传统的原子间潜能与纳米级潜在能量表面 (PESs) 斗争.
- 基于物理学的模型往往缺乏灵活性,与第一原则计算相比,限制了它们对纳米系统的准确性.
- 机器学习 (ML) 为捕捉复杂,尺寸依赖的纳米尺度现象提供了一个有希望的替代方案,而不会影响散装材料的特性.
研究的目的:
- 引入一种新的机器学习工作流程,用于开发高维神经网络 (NN),以获得精确的原子间潜力.
- 使用转移和主动学习策略的组合,同时捕获过渡金属的纳米尺度和散装性能.
- 通过创造多功能和准确的ML训练的潜力来加速材料的发现和设计.
主要方法:
- 开发了一种机器学习工作流程,整合了转移学习和主动学习,以创建高维神经网络 (NN).
- 最初的NN培训使用了现有的高质量的基于物理的模型,用于批量性能,然后再使用第一原则数据进行纳米尺度准确性的再培训.
- 采用稀疏采样,多样化的数据集,涵盖近平衡到非平衡的集群配置,并代改进模型指纹.
主要成果:
- 成功开发了能够捕获10种过渡金属的集群和散装性质的材料不可知性NN.
- 对广泛的第一原则数据进行严格的测试验证了能量,力和质量属性的准确性.
- 机器学习工作流证明了从有限,多样化的数据集中有效学习,改善了传统模型的局限性.
结论:
- 拟议的ML工作流提供了一个强大的方法来创建精确的原子间潜能,弥合纳米尺度和散装材料行为之间的差距.
- 这种方法可以将知识从已建立的模拟转移到新一代的ML训练潜力中,从而加速材料的发现.
- 该工作流的材料不可知性使其在催化,微电子和储能研究中具有广泛的适用性.
相关概念视频
02:58Properties of Transition Metals
25.8K
Transition metals are defined as those elements that have partially filled d orbitals. As shown in Figure 1, the d-block elements in groups 3–12 are transition elements. The f-block elements, also called inner transition metals (the lanthanides and actinides), also meet this criterion because the d orbital is partially occupied before the f orbitals.
25.8K
01:35Long-term Potentiation
55.2K
Long-term potentiation, or LTP, is one of the ways by which synaptic plasticity—changes in the strength of chemical synapses—can occur in the brain. LTP is the process of synaptic strengthening that occurs over time between pre- and postsynaptic neuronal connections. The synaptic strengthening of LTP works in opposition to the synaptic weakening of long-term depression (LTD) and together are the main mechanisms that underlie learning and memory.
55.2K
03:02Colors and Magnetism
11.7K
Color in Coordination Complexes
When atoms or molecules absorb light at the proper frequency, their electrons are excited to higher-energy orbitals. For many main group atoms and molecules, the absorbed photons are in the ultraviolet range of the electromagnetic spectrum, which cannot be detected by the human eye. For coordination compounds, the energy difference between the d orbitals often allows photons in the visible range to be absorbed and emitted, which is seen as colors by the human...
When atoms or molecules absorb light at the proper frequency, their electrons are excited to higher-energy orbitals. For many main group atoms and molecules, the absorbed photons are in the ultraviolet range of the electromagnetic spectrum, which cannot be detected by the human eye. For coordination compounds, the energy difference between the d orbitals often allows photons in the visible range to be absorbed and emitted, which is seen as colors by the human...
11.7K
01:01Neuroplasticity
341
Neuroplasticity reflects the brain's remarkable capacity to adapt and evolve, responding dynamically to learning, experiences, or injury by reorganizing its neural circuitry. This reorganization involves creating new neural connections and refining old ones through a series of biological processes that contribute to the brain's lifelong development and adaptability.
341
02:42Valence Bond Theory
8.5K
Coordination compounds and complexes exhibit different colors, geometries, and magnetic behavior, depending on the metal atom/ion and ligands from which they are composed. In an attempt to explain the bonding and structure of coordination complexes, Linus Pauling proposed the valence bond theory, or VBT, using the concepts of hybridization and the overlapping of the atomic orbitals. According to VBT, the central metal atom or ion (Lewis acid) hybridizes to provide empty orbitals of suitable...
8.5K
02:32Bonding in Metals
47.3K
Metallic bonds are formed between two metal atoms. A simplified model to describe metallic bonding has been developed by Paul Drüde called the “Electron Sea Model”.
47.3K