DALex: Lexicase-Like Selection via Diverse Aggregation

Created by W.Langdon from gp-bibliography.bib Revision:1.7656

@InProceedings{Ni:2024:EuroGP,

• author = "Andrew Ni and Li Ding and Lee Spector",
• editor = "Mario Giacobini and Bing Xue and Luca Manzoni",
• title = "DALex: Lexicase-Like Selection via Diverse Aggregation",
• booktitle = "EuroGP 2024: Proceedings of the 27th European Conference on Genetic Programming",
• year = "2024",
• volume = "14631",
• series = "LNCS",
• publisher = "Springer",
• address = "Aberystwyth",
• month = "3-5 " # apr,
• organisation = "EvoStar, Species",
• keywords = "genetic algorithms, genetic programming",
• pages = "90--107",
• abstract = "Lexicase selection has been shown to provide advantages over other selection algorithms in several areas of evolutionary computation and machine learning. In its standard form, lexicase selection filters a population or other collection based on randomly ordered training cases that are considered one at a time. This iterated filtering process can be time-consuming, particularly in settings with large numbers of training cases, including many symbolic regression and deep learning applications. In this paper, we propose a new method that is nearly equivalent to lexicase selection in terms of the individuals that it selects, but which does so in significantly less time. The new method, called DALex (for Diversely Aggregated Lexicase selection), selects the best individual with respect to a randomly weighted sum of training case errors. This allows us to formulate the core computation required for selection as matrix multiplication instead of recursive loops of comparisons, which in turn allows us to take advantage of optimized and parallel algorithms designed for matrix multiplication for speedup. Furthermore, we show that we can interpolate between the behaviour of lexicase selection and its relaxed variants, such as epsilon and batch lexicase selection, by adjusting a single hyperparameter, named particularity pressure, which represents the importance granted to each individual training case. Results on program synthesis, deep learning, symbolic regression, and learning classifier systems demonstrate that DALex achieves significant speedups over lexicase selection and its relaxed variants while maintaining almost identical problem-solving performance. Under a fixed computational budget, these savings free up resources that can be directed towards increasing population size or the number of generations, enabling the potential for solving more difficult problems.",
• isbn13 = "978-3-031-56957-9",
• DOI = "doi:10.1007/978-3-031-56957-9_6",
• notes = "Part of \cite{Giacobini:2024:GP} EuroGP'2024 held in conjunction with EvoCOP2024, EvoMusArt2024 and EvoApplications2024",

}

Genetic Programming entries for Andrew Ni Li Ding Lee Spector

Citations