Reinforcement learning based metric filtering for evolutionary distance metric learning

被引:2
|
作者
Ali, Bassel [1 ]
Moriyama, Koichi [2 ]
Kalintha, Wasin [1 ]
Numao, Masayuki [3 ]
Fukui, Ken-Ichi [3 ]
机构
[1] Osaka Univ, Grad Sch Informat Sci & Technol, 8-1 Mihogaoka, Osaka 5670047, Japan
[2] Nagoya Inst Technol, Dept Comp Sci, Nagoya, Aichi, Japan
[3] Inst Sci & Ind Res, Osaka, Japan
关键词
Clustering; distance metric learning; feature selection; reinforcement learning;
D O I
10.3233/IDA-194887
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Data collection plays an important role in business agility; data can prove valuable and provide insights for important features. However, conventional data collection methods can be costly and time-consuming. This paper proposes a hybrid system R-EDML that combines a sequential feature selection performed by Reinforcement Learning (RL) with the evolutionary feature prioritization of Evolutionary Distance Metric Learning (EDML) in a clustering process. The goal is to reduce the features while maintaining or increasing the accuracy leading to less time complexity and future data collection time and cost reduction. In this method, features represented by the diagonal elements of EDML matrices are prioritized using a differential evolution algorithm. Further, a selection control strategy using RL is learned by sequentially inserting and evaluating the prioritized elements. The outcome offers the best accuracy R-EDML matrix with the least number of elements. Diagonal R-EDML focusing on the diagonal elements is compared with EDML and conventional feature selection. Full Matrix R-EDML focusing on the diagonal and non-diagonal elements is tested and compared with Information-Theoretic Metric Learning. Moreover, R-EDML policy is tested for each EDML generation and across all generations. Results show a significant decrease in the number of features while maintaining or increasing accuracy.
引用
收藏
页码:1345 / 1364
页数:20
相关论文
共 50 条
  • [41] Distance Metric Learning with Joint Representation Diversification
    Chu, Xu
    Lin, Yang
    Wang, Yasha
    Wang, Xiting
    Yu, Hailong
    Gao, Xin
    Tong, Qi
    25TH AMERICAS CONFERENCE ON INFORMATION SYSTEMS (AMCIS 2019), 2019,
  • [42] Regularized Large Margin Distance Metric Learning
    Li, Ya
    Tian, Xinmei
    Tao, Dacheng
    2016 IEEE 16TH INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), 2016, : 1015 - 1022
  • [43] Toward Certified Robustness of Distance Metric Learning
    Yang, Xiaochen
    Guo, Yiwen
    Dong, Mingzhi
    Xue, Jing-Hao
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, 35 (03) : 3834 - 3844
  • [44] Robust Distance Metric Learning with Auxiliary Knowledge
    Zha, Zheng-Jun
    Mei, Tao
    Wang, Meng
    Wang, Zengfu
    Hua, Xian-Sheng
    21ST INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-09), PROCEEDINGS, 2009, : 1327 - 1332
  • [45] Distance Metric Learning with Joint Representation Diversification
    Chu, Xu
    Lin, Yang
    Wang, Yasha
    Wang, Xiting
    Yu, Hailong
    Gao, Xin
    Tong, Qi
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 119, 2020, 119
  • [46] pyDML: A python library for distance metric learning
    Suárez, Juan Luis
    García, Salvador
    Herrera, Francisco
    Journal of Machine Learning Research, 2020, 21
  • [47] Learning a distance metric from relative comparisons
    Schultz, M
    Joachims, T
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 16, 2004, 16 : 41 - 48
  • [48] Distance Metric Learning with Eigenvalue Fine Tuning
    Wang, Wenquan
    Zhang, Ya
    Hu, Jinglu
    2017 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2017, : 502 - 509
  • [49] Nonlinear Adaptive Distance Metric Learning for Clustering
    Chen, Jianhui
    Zhao, Zheng
    Ye, Jieping
    Liu, Huan
    KDD-2007 PROCEEDINGS OF THE THIRTEENTH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2007, : 123 - 132
  • [50] Distance Metric Learning for Radio Fingerprinting Localization
    Bai, Siqi
    Luo, Yongjie
    Yan, Mingjiang
    Wan, Qun
    EXPERT SYSTEMS WITH APPLICATIONS, 2021, 163 (163)