Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2511.06454 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915985191075840 |
|---|---|
| author | Daniilidis, Aris Corella, Alberto Domínguez Wissgott, Philipp |
| author_facet | Daniilidis, Aris Corella, Alberto Domínguez Wissgott, Philipp |
| contents | We analyze an algorithm for assigning weights prior to scalarization in discrete multi-objective problems arising from data analysis. The algorithm evolves weights (interpreted as the relevance of features) by a replicator-type dynamic on the standard simplex, with update indices computed from a normalized data matrix. We prove that the resulting sequence converges globally to a unique interior equilibrium, yielding non-degenerate limiting weights. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_06454 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Feature weighting for data analysis via evolutionary simulation Daniilidis, Aris Corella, Alberto Domínguez Wissgott, Philipp Optimization and Control Machine Learning We analyze an algorithm for assigning weights prior to scalarization in discrete multi-objective problems arising from data analysis. The algorithm evolves weights (interpreted as the relevance of features) by a replicator-type dynamic on the standard simplex, with update indices computed from a normalized data matrix. We prove that the resulting sequence converges globally to a unique interior equilibrium, yielding non-degenerate limiting weights. |
| title | Feature weighting for data analysis via evolutionary simulation |
| topic | Optimization and Control Machine Learning |
| url | https://arxiv.org/abs/2511.06454 |