On the Conditions for Domain Stability for Machine Learning: a Mathematical Approach
Fuente:
arXiv
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910722415394816 |
|---|---|
| author | Pedroza, Gabriel |
| author_facet | Pedroza, Gabriel |
| contents | This work proposes a mathematical approach that (re)defines a property of Machine Learning models named stability and determines sufficient conditions to validate it. Machine Learning models are represented as functions, and the characteristics in scope depend upon the domain of the function, what allows us to adopt topological and metric spaces theory as a basis. Finally, this work provides some equivalences useful to prove and test stability in Machine Learning models. The results suggest that whenever stability is aligned with the notion of function smoothness, then the stability of Machine Learning models primarily depends upon certain topological, measurable properties of the classification sets within the ML model domain. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_00464 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On the Conditions for Domain Stability for Machine Learning: a Mathematical Approach Pedroza, Gabriel Machine Learning Artificial Intelligence F.4.1; I.2.0 This work proposes a mathematical approach that (re)defines a property of Machine Learning models named stability and determines sufficient conditions to validate it. Machine Learning models are represented as functions, and the characteristics in scope depend upon the domain of the function, what allows us to adopt topological and metric spaces theory as a basis. Finally, this work provides some equivalences useful to prove and test stability in Machine Learning models. The results suggest that whenever stability is aligned with the notion of function smoothness, then the stability of Machine Learning models primarily depends upon certain topological, measurable properties of the classification sets within the ML model domain. |
| title | On the Conditions for Domain Stability for Machine Learning: a Mathematical Approach |
| topic | Machine Learning Artificial Intelligence F.4.1; I.2.0 |
| url | https://arxiv.org/abs/2412.00464 |