Self-Adjusting Evolutionary Algorithms for Multimodal Optimization

Recent theoretical research has shown that self-adjusting and self-adaptive mechanisms can provably outperform static settings in evolutionary algorithms for binary search spaces. However, the vast majority of these studies focuses on unimodal functions which do not require algorithm to flip several bits simultaneously make progress. In fact, existing are designed detect local optima have any obvious benefit cross large Hamming gaps. We suggest a mechanism called stagnation detection be added as module (both with without prior schemes). Added simple (1+1) EA, we prove an expected runtime well-known Jump benchmark corresponds asymptotically optimal parameter setting outperforms other multimodal optimization like heavy-tailed mutation. also investigate context (1+ $$\lambda $$ ) EA. To explore limitations approach, additionally present example where both mechanisms, including detection, help find beneficial mutation rate. Finally, our experimentally.