从预先训练到精确:微调通用原子间潜能,用于精确的催化反应模拟
Jinzhe Ma1, Xiaoyan Fu1, Wenbo Xie1
1School of Physical Science and Technology, ShanghaiTech University, Shanghai 201210, China.
Journal of chemical theory and computation
|February 2, 2026
概括
微调通用机器学习原子间潜力 (uMLIPs) 显著提高了催化反应预测的准确性. 这种方法需要更少的数据,并保留了概括性,使得UMLIP更适用于各种催化系统.
科学领域:
- 计算化学是一种计算化学.
- 材料科学是一种材料科学.
- 机器学习 机器学习
背景情况:
- 万能机器学习原子间潜力 (uMLIPs) 为各种化学系统提供高精度.
- 目前的 uMLIP 难以准确的催化反应和屏障预测.
- 改善MULIP用于反应建模对于催化研究至关重要.
研究的目的:
- 通过微调来提高已建立的 uMLIPs 对于催化反应预测的性能.
- 系统地比较微调与从头开始培训的数据效率和准确性.
- 评估微调对不同任务中的模型概括的影响.
主要方法:
- 评估了两个已建立的 uMLIP.
- 应用了微调策略来提高反应预测的准确性.
- 比较微调与从头开始的训练在各种任务中,如MD模拟,吸附能量和过渡状态搜索.
- 分析了不同训练集大小的表现,并评估了推断概括.
主要成果:
- 微调将反应能量预测的平均绝对误差 (MAE) 从0.38 eV降至0.09 eV.
- 精心调整的模型只需要10%-30%的从头开始培训所需的数据.
- 在微调之后,UMLIP的通用化功能得到了维护.
- 对简单和复杂的任务,包括看不见的元素,都观察到更好的准确性.
结论:
- 微调是一种有效的策略,可以显著提高 uMLIP 对于催化反应预测的准确性.
- 这种方法提高了数据效率,并保持了模型的概括性.
- 微调的 uMLIP 显示出在各种催化反应系统中更广泛应用的巨大潜力.
相关概念视频
Uncertainty in Measurement: Accuracy and Precision
101.2K
Scientists typically make repeated measurements of a quantity to ensure the quality of their findings and to evaluate both the precision and the accuracy of their results. Measurements are said to be precise if they yield very similar results when repeated in the same manner. A measurement is considered accurate if it yields a result that is very close to the true or the accepted value. Precise values agree with each other; accurate values agree with a true value.
101.2K
Fineness of Cement
518
The fineness of cement directly influences the rate of hydration, as the hydration begins at the surface of the cement particles. In addition to hydration, the fineness of cement is vital for various properties of concrete including workability, gypsum requirement, and long-term behavior. The fineness of cement is represented in terms of the specific surface of cement which is typically measured in square meters per kilogram, with several methods available for this determination.
Direct...
Direct...
518
Fineness Modulus
1.5K
The fineness modulus (FM) of aggregate is a numerical index that measures the coarseness or fineness of the particles. It is calculated by adding the cumulative percentages of aggregate retained on each of a specified series of sieves and dividing the sum by 100.
Consider performing sieve analysis on sand through a set of ASTM sieves. The weight of aggregate retained in each sieve and pan placed at the bottom is recorded, as given in Column B of Table 1.
To determine the fineness modulus of...
Consider performing sieve analysis on sand through a set of ASTM sieves. The weight of aggregate retained in each sieve and pan placed at the bottom is recorded, as given in Column B of Table 1.
To determine the fineness modulus of...
1.5K
Turnover Number and Catalytic Efficiency
21.6K
The turnover number of an enzyme is the maximum number of substrate molecules it can transform per unit time. Turnover numbers for most enzymes range from 1 to 1000 molecules per second. Catalase has the known highest turnover number, capable of converting up to 2.8×106 molecules of hydrogen peroxide into water and oxygen per second. Lysozyme has the lowest known turnover number of half a molecule per second.
Chymotrypsin is a pancreatic enzyme that breaks down proteins during digestion....
Chymotrypsin is a pancreatic enzyme that breaks down proteins during digestion....
21.6K
Catalytically Perfect Enzymes
5.1K
The theory of catalytically perfect enzymes was first proposed by W.J. Albery and J. R. Knowles in 1976. These enzymes catalyze biochemical reactions at high-speed. Their catalytic efficiency values range from 108-109 M-1s-1. These enzymes are also called 'diffusion-controlled' as the only rate-limiting step in the catalysis is that of the substrate diffusion into the active site. Examples include triose phosphate isomerase, fumarase, and superoxide dismutase.
Most enzymes...
Most enzymes...
5.1K
Problem-Solving: Tuning of a Guitar String
1.1K
In the case of stringed instruments like the guitar, the elastic property that determines the speed of the sound produced is its linear mass density or the mass per unit length. This is simply called the linear density. If the string's linear density is constant along the string, then the linear density is simply the total mass divided by the total length.
The string's wave speed can be regulated by varying the linear density. Tension is the other property that determines the speed of...
The string's wave speed can be regulated by varying the linear density. Tension is the other property that determines the speed of...
1.1K


