A knowledge-based approach to supervised incremental learning

淡江大學機構典藏 > 工學院 > 資訊工程學系暨研究所 > 會議論文 > Item 987654321/37525

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/37525

题名:	A knowledge-based approach to supervised incremental learning
作者:	Fu, Li-min;Hsu, Hui-huang;Principe, Jose C.
贡献者:	淡江大學資訊工程學系
日期:	1994-06-27
上传时间:	2010-04-15 09:45:51 (UTC+8)
出版者:	Institute of Electrical and Electronics Engineers (IEEE)
摘要:	How to learn new knowledge without forgetting old knowledge is a key issue in designing an incremental-learning neural network. In this paper, we present a rule-based connectionist approach in which old knowledge is preserved by bounding weight modifications. In addition, some heuristics are developed for avoiding overtraining of the network and adding new hidden units. The feasibility of this approach is demonstrated for classification problems including the iris and the promoter domains.
關聯:	Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on, vol.3, pp.1793-1798
DOI:	10.1109/ICNN.1994.374428
显示于类别:	[資訊工程學系暨研究所] 會議論文

文件中的档案:

档案	描述	大小	格式	浏览次数
078031901X_3p1793-1798.pdf		393Kb	Adobe PDF	627	检视/开启
index.html		0Kb	HTML	283	检视/开启

在機構典藏中所有的数据项都受到原著作权保护.

TAIR相关文章