本文借助于输入的函数扩展思想，将其引入到一般类型的联想存储器模型，通过噪声输入模式来优化所提出的模型，推广了一类联想存储器模型，实验结果表明最小平方函数链式联想存储器（ＦＬＡＭ）比之ＫＯＨＯＮＥＮ模型、ＭＵＲＡＫＡＭＩ模型在性能上优越．

This paper introduces the idea of input functional expansion into a type of general associative memories, and proposes a new least squares functional link associative memory model, the performance of the model is optimized through noise input patterns.The experimental, results show the associative effect of the proposed model is better than those of the Kohonen s model and the Murakami s model.

本文部分研究获“攀登计划”资助.