Symbolic Regression via Control Variable Genetic Programming

被引:2
|
作者
Jiang, Nan [1 ]
Xue, Yexiang [1 ]
机构
[1] Purdue Univ, Dept Comp Sci, W Lafayette, IN 47907 USA
来源
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES: RESEARCH TRACK, ECML PKDD 2023, PT IV | 2023年 / 14172卷
关键词
Control Variable Experiment; Symbolic Regression; ALGORITHMS; DISCOVERY;
D O I
10.1007/978-3-031-43421-1_11
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Learning symbolic expressions directly from experiment data is a vital step in AI-driven scientific discovery. Nevertheless, state-of-the-art approaches are limited to learning simple expressions. Regressing expressions involving many independent variables still remain out of reach. Motivated by the control variable experiments widely utilized in science, we propose Control Variable Genetic Programming (CVGP) for symbolic regression over many independent variables. CVGP expedites symbolic expression discovery via customized experiment design, rather than learning from a fixed dataset collected a priori. CVGP starts by fitting simple expressions involving a small set of independent variables using genetic programming, under controlled experiments where other variables are held as constants. It then extends expressions learned in previous generations by adding new independent variables, using new control variable experiments in which these variables are allowed to vary. Theoretically, we show CVGP as an incremental building approach can yield an exponential reduction in the search space when learning a class of expressions. Experimentally, CVGP outperforms several baselines in learning symbolic expressions involving multiple independent variables.
引用
收藏
页码:178 / 195
页数:18
相关论文
共 50 条
  • [1] Racing Control Variable Genetic Programming for Symbolic Regression
    Jiang, Nan
    Xue, Yexiang
    THIRTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 11, 2024, : 12901 - 12909
  • [2] Variable neighborhood programming for symbolic regression
    Souhir Elleuch
    Bassem Jarboui
    Nenad Mladenovic
    Jun Pei
    Optimization Letters, 2022, 16 : 191 - 210
  • [3] Variable neighborhood programming for symbolic regression
    Elleuch, Souhir
    Jarboui, Bassem
    Mladenovic, Nenad
    Pei, Jun
    OPTIMIZATION LETTERS, 2022, 16 (01) : 191 - 210
  • [4] Taylor Genetic Programming for Symbolic Regression
    He, Baihe
    Lu, Qiang
    Yang, Qingyun
    Luo, Jake
    Wang, Zhiguang
    PROCEEDINGS OF THE 2022 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO'22), 2022, : 946 - 954
  • [5] Statistical genetic programming for symbolic regression
    Haeri, Maryam Amir
    Ebadzadeh, Mohammad Mehdi
    Folino, Gianluigi
    APPLIED SOFT COMPUTING, 2017, 60 : 447 - 469
  • [6] Compositional Genetic Programming for Symbolic Regression
    Krawiec, Krzysztof
    Kossinski, Dominik
    PROCEEDINGS OF THE 2022 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE COMPANION, GECCO 2022, 2022, : 570 - 573
  • [7] The Inefficiency of Genetic Programming for Symbolic Regression
    Kronberger, Gabriel
    de Franca, Fabricio Olivetti
    Desmond, Harry
    Bartlett, Deaglan J.
    Kammerer, Lukas
    PARALLEL PROBLEM SOLVING FROM NATURE-PPSN XVIII, PPSN 2024, PT I, 2024, 15148 : 273 - 289
  • [8] Lifetime Adaptation in Genetic Programming for the Symbolic Regression
    Merta, Jan
    Brandejsky, Tomas
    COMPUTATIONAL STATISTICS AND MATHEMATICAL MODELING METHODS IN INTELLIGENT SYSTEMS, VOL. 2, 2019, 1047 : 339 - 346
  • [9] Symbol Graph Genetic Programming for Symbolic Regression
    Song, Jinglu
    Lu, Qiang
    Tian, Bozhou
    Zhang, Jingwen
    Luo, Jake
    Wang, Zhiguang
    PARALLEL PROBLEM SOLVING FROM NATURE-PPSN XVIII, PPSN 2024, PT I, 2024, 15148 : 221 - 237
  • [10] Genetic programming with separability detection for symbolic regression
    Liu, Wei-Li
    Yang, Jiaquan
    Zhong, Jinghui
    Wang, Shibin
    COMPLEX & INTELLIGENT SYSTEMS, 2021, 7 (03) : 1185 - 1194