A Unified Framework for Deep Symbolic Regression

Created by W.Langdon from gp-bibliography.bib Revision:1.7650

@InProceedings{DBLP:conf/nips/LandajuelaLYGSA22,

• author = "Mikel Landajuela and Chak Shing Lee and Jiachen Yang and Ruben Glatt and Claudio P. Santiago and Ignacio Aravena and Terrell Nathan Mundhenk and Garrett Mulcahy and Brenden K. Petersen",
• title = "A Unified Framework for Deep Symbolic Regression",
• booktitle = "Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems, NeurIPS 2022",
• year = "2022",
• editor = "Sanmi Koyejo and S. Mohamed and A. Agarwal and Danielle Belgrave and K. Cho and A. Oh",
• pages = "33985--33998",
• address = "New Orleans, USA",
• month = nov # " 28 - " # dec # " 9",
• publisher = "Curran Associates, Inc.",
• keywords = "genetic algorithms, genetic programming",
• timestamp = "Mon, 08 Jan 2024 16:31:35 +0100",
• biburl = "https://dblp.org/rec/conf/nips/LandajuelaLYGSA22.bib",
• bibsource = "dblp computer science bibliography, https://dblp.org",
• URL = "http://papers.nips.cc/paper_files/paper/2022/hash/dbca58f35bddc6e4003b2dd80e42f838-Abstract-Conference.html",
• URL = "https://proceedings.neurips.cc/paper_files/paper/2022/hash/dbca58f35bddc6e4003b2dd80e42f838-Abstract-Conference.html",
• URL = "https://proceedings.neurips.cc/paper_files/paper/2022/file/dbca58f35bddc6e4003b2dd80e42f838-Paper-Conference.pdf",
• code_url = "https://github.com/cavalab/srbench",
• size = "14 pages",
• abstract = "The last few years have witnessed a surge in methods for symbolic regression, from advances in traditional evolutionary approaches to novel deep learning-based systems. Individual works typically focus on advancing the state-of-the-art for one particular class of solution strategies, and there have been few attempts to investigate the benefits of hybridizing or integrating multiple strategies. In this work, we identify five classes of symbolic regression solution strategies, recursive problem simplification, neural-guided search, large-scale pre-training, genetic programming, and linear models, and propose a strategy to hybridize them into a single modular, unified symbolic regression framework. Based on empirical evaluation using SRBench, a new community tool for benchmarking symbolic regression methods, our unified framework achieves state-of-the-art performance in its ability to (1) symbolically recover analytical expressions, (2) fit datasets with high accuracy, and (3) balance accuracy-complexity trade-offs, across 252 ground-truth and black-box benchmark problems, in both noiseless settings and across various noise levels. Finally, we provide practical use case-based guidance for constructing hybrid symbolic regression algorithms, supported by extensive, combinatorial ablation studies.",
• notes = "also known as \cite{NEURIPS2022_dbca58f3}",

}

Genetic Programming entries for Mikel Landajuela Chak Shing Lee Jiachen Yang Ruben Glatt Claudio P Santiago Ignacio Aravena T Nathan Mundhenk Garrett Mulcahy Brenden Kyle Petersen

Citations