English

Feature weighting for data analysis via evolutionary simulation

Optimization and Control 2026-05-08 v2 Machine Learning

Abstract

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.

Keywords

Cite

@article{arxiv.2511.06454,
  title  = {Feature weighting for data analysis via evolutionary simulation},
  author = {Aris Daniilidis and Alberto Domínguez Corella and Philipp Wissgott},
  journal= {arXiv preprint arXiv:2511.06454},
  year   = {2026}
}
R2 v1 2026-07-01T07:28:28.071Z