Adaptive sampling for active learning with genetic programming

被引:5
|
作者
Ben Hamida, Sana [1 ]
Hmida, Hmida [1 ,2 ]
Borgi, Amel [3 ,4 ]
Rukoz, Marta [1 ,5 ]
机构
[1] Univ Paris 09, CNRS, LAMSADE, PSL Res Univ,UMR 7243, F-75016 Paris, France
[2] Univ Tunis El Manar, Fac Sci Tunis, LR11ES14 LIPAH, Tunis 2092, Tunisia
[3] Univ Tunis El Manar, Inst Super Informat, Tunis 2092, Tunisia
[4] Fac Sci Tunis, LR11ES14 LIPAH, Tunis 2092, Tunisia
[5] Univ Paris Nanterre, F-92001 Nanterre, France
来源
COGNITIVE SYSTEMS RESEARCH | 2021年 / 65卷 / 65期
关键词
Genetic programming; Machine learning; Active learning; Training data sampling; Adaptive sampling; Sampling frequency control; SUBSET-SELECTION;
D O I
10.1016/j.cogsys.2020.08.008
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Active learning is a machine learning paradigm allowing to decide which inputs to use for training. It is introduced to Genetic Programming (GP) essentially thanks to the dynamic data sampling, used to address some known issues such as the computational cost, the over-fitting problem and the imbalanced databases. The traditional dynamic sampling for GP gives to the algorithm a new sample periodically, often each generation, without considering the state of the evolution. In so doing, individuals do not have enough time to extract the hidden knowledge. An alternative approach is to use some information about the learning state to adapt the periodicity of the training data change. In this work, we propose an adaptive sampling strategy for classification tasks based on the state of solved fitness cases throughout learning. It is a flexible approach that could be applied with any dynamic sampling. We implemented some sampling algorithms extended with dynamic and adaptive controlling re-sampling frequency. We experimented them to solve the KDD intrusion detection and the Adult incomes prediction problems with GP. The experimental study demonstrates how the sampling frequency control preserves the power of dynamic sampling with possible improvements in learning time and quality. We also demonstrate that adaptive sampling can be an alternative to multi-level sampling. This work opens many new relevant extension paths. (c) 2020 Elsevier B.V. All rights reserved.
引用
收藏
页码:23 / 39
页数:17
相关论文
共 50 条
  • [1] A new adaptive sampling approach for Genetic Programming
    Hmida, Hmida
    Ben Hamida, Sana
    Borgi, Amel
    Rukoz, Marta
    2019 THIRD INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTING IN DATA SCIENCES (ICDS 2019), 2019,
  • [2] Adaptive Sampling of Biomedical Images with Cartesian Genetic Programming
    Lavinas, Yuri
    Haut, Nathan
    Punch, William
    Banzhaf, Wolfgang
    Cussat-Blanc, Sylvain
    PARALLEL PROBLEM SOLVING FROM NATURE-PPSN XVIII, PPSN 2024, PT I, 2024, 15148 : 256 - 272
  • [3] A Study on Genetic Programming with Layered Learning and Incremental Sampling
    Nguyen Thi Hien
    Nguyen Xuan Hoai
    McKay, Bob
    2011 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2011, : 1179 - 1185
  • [4] Active Learning Genetic Programming for Record Deduplication
    de Freitas, Junio
    Pappa, Gisele L.
    da Silva, Altigran S.
    Goncalves, Marcos A.
    Moura, Edleno
    Veloso, Adriano
    Laender, Alberto H. F.
    de Carvalho, Moises G.
    2010 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2010,
  • [5] Active Sampling for Dynamic Job Shop Scheduling using Genetic Programming
    Karunakaran, Deepak
    Mei, Yi
    Chen, Gang
    Zhang, Mengjie
    2019 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2019, : 434 - 441
  • [6] On sampling error in genetic programming
    Dirk Schweim
    David Wittenberg
    Franz Rothlauf
    Natural Computing, 2022, 21 : 173 - 186
  • [7] On sampling error in genetic programming
    Schweim, Dirk
    Wittenberg, David
    Rothlauf, Franz
    NATURAL COMPUTING, 2022, 21 (02) : 173 - 186
  • [8] Active learning of expressive linkage rules using genetic programming
    Isele, Robert
    Bizer, Christian
    JOURNAL OF WEB SEMANTICS, 2013, 23 : 2 - 15
  • [9] Regex-based Entity Extraction with Active Learning and Genetic Programming
    Bartoli, Alberto
    De Lorenzo, Andrea
    Medvet, Eric
    Tarlao, Fabiano
    APPLIED COMPUTING REVIEW, 2016, 16 (02): : 7 - 15
  • [10] Active Learning Informs Symbolic Regression Model Development in Genetic Programming
    Haut, Nathan
    Punch, Bill
    Banzhaf, Wolfgang
    PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE COMPANION, GECCO 2023 COMPANION, 2023, : 587 - 590