Extremely large-scale optimization problem-solving using robust and scalable evolutionary computation

我们开发了一系列进化算法，如BOA- md （BOA的扩展，引入混合分布），将BOA与局部搜索相结合的BHCS，用于MINLP（混合整数非线性规划）的ARGA和BRGA。我们还在多核心架构和云计算环境中开发了大规模并行进化算法。作为对现实问题的应用，我们将进化算法应用于云计算环境下的资源优化分配问题，作为一个MINLP问题，将进化算法应用于配体对接问题，自动发现药物的有前途的结构，等等，以显示我们方法的有效性。

We have developed a series of evolutionary algorithms such as BOA-MD, an extension of BOA introducing mixture distributions, BHCS by combining BOA with local search, ARGA and BRGA for MINLP (Mixed-Interger Non-Linear Programming). We also developed large-scale parallel evolutionary algorithms in a many core architecture and cloud computing environment.As applications to real-world problems, we applied evolutionary algorithms to optimal resource allocation problem in cloud computing environment as a MINLP problem, de novo ligand docking problems to find promising structures of medicines automatically, and so on, to show the effectiveness of our approach.