ML CSIG: Difference between revisions

From VASP Wiki
No edit summary
No edit summary
 
(7 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
{{DISPLAYTITLE:ML_CSIG}}
{{TAGDEF|ML_CSIG|[real]|<math>0.4</math>}}
{{TAGDEF|ML_CSIG|[real]|<math>0.4</math>}}


Description: Parameter used in the automatic determination of threshold {{TAG|ML_CTIFOR}} for Bayesian error estimation in the machine learning force field method.
Description: Parameter used in the automatic determination of threshold {{TAG|ML_CTIFOR}} for error estimation in the machine learning force field method.
----
----
For details please read entry {{TAG|ML_ICRITERIA}} first. The parameter {{TAG|ML_CTIFOR}} is only updated, if the standard error of the collected Bayesian errors is below {{TAG|ML_CSIG}} times the mean of the collected Bayesian errors.
The usage of this tag in combination with the learning algorithms is described here: [[Machine learning force field calculations: Basics#Threshold for error of forces|here]].
 
The standard error of the history of maximum estimated errors of the forces ({{TAG|ML_MHIS}}) and it's slope must be below {{TAG|ML_CSIG}} and {{TAG|ML_CSLOPE}} so that an update of the threshold for the maximum estimated error of forces {{TAG|ML_CTIFOR}} can take place.  


== Related Tags and Sections ==
== Related tags and articles ==
{{TAG|ML_LMLFF}}, {{TAG|ML_ICRITERIA}}, {{TAG|ML_CSLOPE}}, {{TAG|ML_MHIS}}, {{TAG|ML_CTIFOR}}
{{TAG|ML_LMLFF}}, {{TAG|ML_ICRITERIA}}, {{TAG|ML_CSLOPE}}, {{TAG|ML_MHIS}}, {{TAG|ML_CTIFOR}}, {{TAG|ML_CX}}


{{sc|ML_CSIG|Examples|Examples that use this tag}}
{{sc|ML_CSIG|Examples|Examples that use this tag}}
----
----


[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category: Alpha]]
[[Category:INCAR tag]][[Category:Machine-learned force fields]]

Latest revision as of 12:30, 12 June 2024

ML_CSIG = [real]
Default: ML_CSIG =  

Description: Parameter used in the automatic determination of threshold ML_CTIFOR for error estimation in the machine learning force field method.


The usage of this tag in combination with the learning algorithms is described here: here.

The standard error of the history of maximum estimated errors of the forces (ML_MHIS) and it's slope must be below ML_CSIG and ML_CSLOPE so that an update of the threshold for the maximum estimated error of forces ML_CTIFOR can take place.

Related tags and articles

ML_LMLFF, ML_ICRITERIA, ML_CSLOPE, ML_MHIS, ML_CTIFOR, ML_CX

Examples that use this tag