A skewness-based stopping rule estimates the required number of algorithm runs online and, in large COCO benchmarks, achieves 82 to 95 percent estimation accuracy while saving roughly 50 percent of runs, with a 5 to 25 percent error rate.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Estimation of the Number of Algorithm Runs in Stochastic Optimization
A skewness-based stopping rule estimates the required number of algorithm runs online and, in large COCO benchmarks, achieves 82 to 95 percent estimation accuracy while saving roughly 50 percent of runs, with a 5 to 25 percent error rate.