首页> 外文期刊>Journal of Computer and System Sciences >The Learnability of Unions of Two Rectangles in the Two- Dimensional Discretized Space
【24h】

The Learnability of Unions of Two Rectangles in the Two- Dimensional Discretized Space

机译:二维离散空间中两个矩形并集的可学习性

获取原文
获取原文并翻译 | 示例
       

摘要

We study the problem of properly learning unions of two axis-parallel rec- tangles over the domain {0,n-1}~2 in the on-line model with equivalence queries. When only O(log n) equivalence queries are allowed, this problem is one of the five interesting open problems proposed by W. Maass and G. Turan (Mach. Learning 14, 1994, 251--269), regarding learning geometric concepts. In this paper, we design an efficient algorithm that properly learns unions of two rectangles over the domain { 0, n -- 1 }~2 using O(log~2 n) equiv- alence queries.
机译:我们研究了在具有等价查询的在线模型中,在域{0,n-1}〜2上正确学习两个轴平行矩形的并集的问题。当仅允许O(log n)等价查询时,此问题是W.Maass和G.Turan(Mach.Learning 14,1994,251--269)提出的有关学习几何概念的五个有趣的开放问题之一。在本文中,我们设计了一种有效的算法,该算法使用O(log〜2 n)等价查询正确学习域{0,n-1}〜2上两个矩形的并集。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号