噪声环境下 NSGA-II 的鲁棒性分析
Analysing the Robustness of NSGA-II under Noise
AI总结:
本文通过后验噪声模型分析 NSGA-II 与 GSEMO,证明在 p<1/2 时 NSGA-II 可保留有效个体并在 LeadingOnesTrailingZeroes 上优于 GSEMO,且 p=1/2 存在多项式到指数时间的相变。
摘要:
运行时间分析已针对 (1+1) EA 等简单进化算法,以及其在进化多目标优化(EMO)中的对应算法 GSEMO 的效率得出了许多结果。最近,对著名且被大量引用的 EMO 算法 NSGA-II 的首批运行时间分析开始出现,证明了拥有数千种应用的实用算法也可以接受严格分析。然而,这些结果仅表明 NSGA-II 具有与 GSEMO 相同的性能保证,尚不清楚 NSGA-II 如何以及何时能够优于 GSEMO。我们在噪声优化中研究这一问题,并考虑一种噪声模型:在每次评估时,以某个常数概率 p 向所有目标添加大量后验噪声。我们证明,GSEMO 在每个含噪适应度函数上都会严重失效,因为它往往会不加区分地删除种群的大部分内容。相比之下,当 p<1/2 时,NSGA-II 能够在 \ extsc{LeadingOnesTrailingZeroes} 上高效处理噪声,因为该算法即使在噪声存在时也能保留有用的搜索点。我们识别出 p=1/2 处的相变:覆盖 Pareto 前沿的期望时间在此由多项式变为指数级。据我们所知,这是 NSGA-II 能够优于 GSEMO 的首个证明,也是 NSGA-II 在噪声优化中的首个运行时间分析。
英文摘要:
Runtime analysis has produced many results on the efficiency of simple evolutionary algorithms like the (1+1) EA, and its analogue called GSEMO in evolutionary multiobjective optimisation (EMO). Recently, the first runtime analyses of the famous and highly cited EMO algorithm NSGA-II have emerged, demonstrating that practical algorithms with thousands of applications can be rigorously analysed. However, these results only show that NSGA-II has the same performance guarantees as GSEMO and it is unclear how and when NSGA-II can outperform GSEMO. We study this question in noisy optimisation and consider a noise model that adds large amounts of posterior noise to all objectives with some constant probability $p$ per evaluation. We show that GSEMO fails badly on every noisy fitness function as it tends to remove large parts of the population indiscriminately. In contrast, NSGA-II is able to handle the noise efficiently on \textsc{LeadingOnesTrailingZeroes} when $p<1/2$, as the algorithm is able to preserve useful search points even in the presence of noise. We identify a phase transition at $p=1/2$ where the expected time to cover the Pareto front changes from polynomial to exponential. To our knowledge, this is the first proof that NSGA-II can outperform GSEMO and the first runtime analysis of NSGA-II in noisy optimisation.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。