On-line learning algorithm for organizing data based on generalized entropy

基于广义熵的数据组织在线学习算法

• 批准号: 13480074
• 负责人: MARUOKA Akira
• 金额: $ 10.05万
• 依托单位: Tohoku University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 2001
• 资助国家: 日本
• 起止时间: 2001 至 2002
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-13480074/
• 关键词: boosting; learning from examples; on-line learning algorithm; prunning; decision tree boosting; generalized entropy; learning curve; over-fitting; マージン; サポートベクトルマシーン; 勾配傾斜法; ランダムプロジェクション; 次元圧縮; m-限定独立

The amount of data collected from various fields is growing exponentially and the task of exracting useful information from data is becoming more and more difficult accordingly. To overcome the difficulty that comes from the limitation on computational resources, we investigate various methodologies of exracting useful information from huge data by organizing consice data based on pseudo-entropy function. The results we obtained includes the following algorithms : top-down decision tree learning algorithm based on information based boosting ; an algorithm to obtain the nearly best prunnig of a decision trre ; an algorithm to learn monotone log-term DNF formulas under uniform distribution.

从各个领域收集的数据量呈指数级增长，从数据中提取有用信息的任务相应地变得越来越困难。为了克服计算资源有限所带来的困难，本文研究了基于伪熵函数对海量数据进行组织，从海量数据中提取有用信息的各种方法。我们得到的结果包括以下算法：基于信息提升的自顶向下决策树学习算法;决策树的近最佳剪枝算法;均匀分布下单调对数项DNF公式的学习算法。

项目成果

Kazuyuki Amano, Akira Maruoka: "On learning monotone Boolean functions under the uniform distribution"Proc. of the 13th International Conference on Algorithmic Learning. Vol.2533. 57-68 (2002)

Kazuyuki Amano、Akira Maruoka：“论学习均匀分布下的单调布尔函数”Proc。

Kazuyuki Amano, Akira Maruoka: "On the negation-limited circuit complexity of merging"Discrete Applied Mathematics. 126. 3-8 (2003)

Kazuyuki Amano、Akira Maruoka：“论合并的否定限制电路复杂性”离散应用数学。

Kazuyuki Amano, Akira Mruoka: "On the negation-limited circuit complexity of merging"Discrete Applied Mathematics. Vol.126. 3-8 (2003)

Kazuyuki Amano、Akira Mruoka：“论合并的否定限制电路复杂性”离散应用数学。

Akira Maruoka, Eiji Takimoto: "Algorithmic learning theory"MARCEL DEKKER, INC.. 43 (2002)

Akira Maruoka、Eiji Takimoto：“算法学习理论”MARCEL DEKKER, INC.. 43 (2002)

Eiji Takimoto and Manfred Warmuth: "Path kernels and multiplicative updates"Lecture Notes in Artificial Intelligence. 2375. 74-89 (2002)

Eiji Takimoto 和 Manfred Warmuth：“路径内核和乘法更新”人工智能讲义。