一种基于增量学习的典型样本选取方法

An Algorithm for Representative Sampling Based on Incremental Learning

提出了一种基于增量学习的典型样本选取方法,旨在大样本情况下获取具有学习任务所需的充分信息量且规模最小的训练集用于神经网络建模。仿真结果表明,该方法有利于缩短训练时间和提高神经网络泛化能力,从而具有很好的实用性。

An algorithm for representative sampling based on incremental learning is presented in this paper,in order to obtain smallest training set from larger data sets,which contain enough information for learning the task in usage of neural network modeling.The simulation results show that selecting representative sampling is in favor of shorting the training time and improving the generalization ability.