摘要
提出示例学习的一种几何理论，即扩张图方法。应用该方法求得了最优覆盖问题的近似解。该方法直观、清晰，并具有渐近式学习功能。
A new geometric theory of learning from examples, the extension graph (EG) ap- proach, is presented and has been used to give an approximate solution to the minimal covers (MCV)problem. EG, a straightforward and distinct approach, has also the capability of incremental learning.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
1998年第1期65-67,72,共4页
Journal of Harbin Institute of Technology
关键词
示例学习
扩张图
渐近式学习
机器学习
专家系统
Learning from examples: extension graph
incremental learning: concept acquisition
