GeM-EA uses bi-level meta-learning for surrogate initialization and generative replay in a multi-island evolutionary strategy to achieve faster adaptation and robustness in streaming data-driven optimization under concept drift.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GeM-EA: A Generative and Meta-learning Enhanced Evolutionary Algorithm for Streaming Data-Driven Optimization
GeM-EA uses bi-level meta-learning for surrogate initialization and generative replay in a multi-island evolutionary strategy to achieve faster adaptation and robustness in streaming data-driven optimization under concept drift.