On-the-fly simplification of genetic programming models

被引:4
|
作者
Javed, Noman [1 ]
Gobet, Fernand [1 ]
机构
[1] London Sch Econ & Polit Sci, London, England
来源
36TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, SAC 2021 | 2021年
基金
欧洲研究理事会;
关键词
Evolutionary Computing; Genetic Programming; Simplification; BLOAT;
D O I
10.1145/3412841.3441926
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The last decade has seen amazing performance improvements in deep learning. However, the black-box nature of this approach makes it difficult to provide explanations of the generated models. In some fields such as psychology and neuroscience, this limitation in explainability and interpretability is an important issue. Approaches such as genetic programming are well positioned to take the lead in these fields because of their inherent white box nature. Genetic programming, inspired by Darwinian theory of evolution, is a population-based search technique capable of exploring a high-dimensional search space intelligently and discovering multiple solutions. However, it is prone to generate very large solutions, a phenomenon often called "bloat". The bloated solutions are not easily understandable. In this paper, we propose two techniques for simplifying the generated models. Both techniques are tested by generating models for a well-known psychology experiment. The validity of these techniques is further tested by applying them to a symbolic regression problem. Several population dynamics are studied to make sure that these techniques are not compromising diversity - an important measure for finding better solutions. The results indicate that the two techniques can be both applied independently and simultaneously and that they are capable of finding solutions at par with those generated by the standard GP algorithm - but with significantly reduced program size. There was no loss in diversity nor reduction in overall fitness. In fact, in some experiments, the two techniques even improved fitness.
引用
收藏
页码:464 / 471
页数:8
相关论文
共 50 条
  • [31] On-the-fly Table Generation
    Zhang, Shuo
    Balog, Krisztian
    ACM/SIGIR PROCEEDINGS 2018, 2018, : 595 - 604
  • [32] Evolution on-the-fly with Paradigm
    Groenewegen, Luuk
    de Vink, Erik
    COORDINATION MODELS AND LANGUAGES, PROCEEDINGS, 2006, 4038 : 97 - 112
  • [33] On-the-fly reconfigurable logic
    Rajagopalan, K
    Phillips, B
    Abbott, D
    SMART STRUCTURES, DEVICES, AND SYSTEMS II, PT 1 AND 2, 2005, 5649 : 101 - 109
  • [34] Materials Design On-the-Fly
    Cerqueira, Tiago F. T.
    Sarmiento-Perez, Rafael
    Amsler, Maximilian
    Nogueira, F.
    Botti, Silvana
    Marques, Miguel A. L.
    JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2015, 11 (08) : 3955 - 3960
  • [35] On-the-fly pipeline parallelism
    Lee, I.-T.A.
    Leiserson, C.E.
    Schardl, T.B.
    Zhang, Z.
    Sukha, J.
    ACM Transactions on Parallel Computing, 2015, 2 (03)
  • [36] Variational On-the-Fly Personalization
    Kim, Jangho
    Lee, Jun-Tae
    Chang, Simyung
    Kwak, Nojun
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022,
  • [37] On-the-fly pipelined convolver
    Marino, F
    ELECTRONICS LETTERS, 1998, 34 (12) : 1198 - 1200
  • [38] An LSI implementation of an adaptive genetic algorithm with on-the-fly crossover operator selection
    Wakabayashi, S
    Koide, T
    Toshine, N
    Goto, M
    Nakayama, Y
    Hatta, K
    PROCEEDINGS OF ASP-DAC '99: ASIA AND SOUTH PACIFIC DESIGN AUTOMATION CONFERENCE 1999, 1999, : 37 - 40
  • [39] On-The-Fly Capacity Planning
    Mitchell, Nick
    Sweeney, Peter F.
    ACM SIGPLAN NOTICES, 2013, 48 (10) : 849 - 866
  • [40] On-the-fly calibration at STScI
    Lubow, S
    Pollizzi, J
    ASTRONOMICAL DATA ANALYSIS SOFTWARE AND SYSTEMS VIII, 1999, 172 : 187 - 190