Numerosity Reduction for Resource Constrained Learning (Preprint)

スポンサーリンク

概要

• 論文の詳細を見る

• When coupling data mining (DM) and learning agents, one of the crucial challenges is the need for the Knowledge Extraction (KE) process to be lightweight enough so that even resource (e.g., memory, CPU etc.) constrained agents are able to extract knowledge. We propose the Stratified Ordered Selection (SOS) method for achieving lightweight KE using dynamic numerosity reduction of training examples. SOS allows for agents to retrieve different-sized training subsets based on available resources. The method employs ranking-based subset selection using a novel Level Order (LO) ranking scheme. We show representativeness of subsets selected using the proposed method, its noise tolerance nature and ability to preserve KE performance over different reduction levels. When compared to subset selection methods of the same category, the proposed method offers the best trade-off between cost, reduction and the ability to preserve performance.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.21(2013) No.2 (online)DOI http://dx.doi.org/10.2197/ipsjjip.21.329------------------------------

• 2013-04-15

著者

• Gen Kitagata

  Graduate School of Information Sciences, Tohoku University,Research Institute of Electrical Communic

• Gen Kitagata

  Research Institute of Electrical Communication Tohoku University Graduate School of Information Scie

• Tetsuo Kinoshita

  Graduate School Of Information Sciences Tohoku University | Research Institute Of Electrical Communi

• Hideyuki Takahashi

  Research Institute Of Electrical Communication Tohoku University

• Hideyuki Takahashi

  Graduate School of Information Sciences, Tohoku University|Research Institute of Electrical Communication, Tohoku University

• Khamisi Kalegele

  Graduate School of Information Sciences, Tohoku University

• Johan Sveholm

  Research Institute of Electrical Communication, Tohoku University

• Kazuto Sasai

  Graduate School of Information Sciences, Tohoku University|Research Institute of Electrical Communication, Tohoku University

• Gen Kitagata

  Graduate School of Information Sciences, Tohoku University|Research Institute of Electrical Communication, Tohoku University

• Tetsuo Kinoshita

  Graduate School of Information Sciences, Tohoku University|Research Institute of Electrical Communication, Tohoku University

関連論文

• Design of Adaptive Communication Mechanism for Ubiquitous Multiagent Systems
• Proposal and Performance Evaluation of Hash-based Authentication for P2P Network
• A New QoS Routing Scheme Based on Bandwidth Consumption for MANETs
• Numerosity Reduction for Resource Constrained Learning (Preprint)
• Design and Evaluation of Energy-consumption-aware Evolutional Agent System for Portable Devices (Preprint)

スポンサーリンク