改进人工蜂群算法优化ELM分类模型

Improved Artificial Bee Colony optimize ELM classification model

摘要:

针对极限学习机(Extreme Learning Machine,ELM)参数优化问题,提出改进人工蜂群算法(Improvement Artificial bee colony, IABC)优化ELM分类模型。算法采用解更新策略池代替固定不变的更新策略,将邻域搜索自适应化；优化侦察蜂搜索方式,利用Kent映射产生均匀性更优的初始随机数序列。在分类数据集中,将IABC-ELM分类模型同ELM、PSO-ELM分类模型进行对比实验。实验中,IABC-ELM模型取得了最佳的分类结果,得到了最低的输出权重范数。结果表明,IABC-ELM模型分类效果显著优于对比模型,证实了IABC算法优化ELM分类模型的有效性和优越性。

Abstract:

Due to the drawbacks of parameter optimization in Extreme learning Machine, an Improved Artificial Bee Colony was proposed to optimize ELM classification model. In the IABC algorithm, the solution update strategy pool was used to replace the fixed update strategy; Optimized the search mode of scouts; The initial random number sequence generated by Kent mapping for better uniformity. In the classification data set, IABC-ELM classification model was compared with the ELM and PSO-ELM classification model. IABC-ELM model obtains the best classification result and the lowest output weight norm. The results show that the classification performance of IABC-ELM model is significantly better than that of the contrast model. Confirmed the validity and superiority of the IABC algorithm to optimize the ELM classification model.