Large Language Model-based Test Case Generation for GP Agents

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@InProceedings{jorgensen:2024:GECCO,

• author = "Steven Jorgensen and Giorgia Nadizar and Gloria Pietropolli and Luca Manzoni and Eric Medvet and Una-May O'Reilly and Erik Hemberg",
• title = "Large Language Model-based Test Case Generation for {GP} Agents",
• booktitle = "Proceedings of the 2024 Genetic and Evolutionary Computation Conference",
• year = "2024",
• editor = "Ting Hu and Aniko Ekart and Julia Handl and Xiaodong Li and Markus Wagner and Mario Garza-Fabre and Kate Smith-Miles and Richard Allmendinger and Ying Bi and Grant Dick and Amir H Gandomi and Marcella Scoczynski Ribeiro Martins and Hirad Assimi and Nadarajen Veerapen and Yuan Sun and Mario Andres Munyoz and Ahmed Kheiri and Nguyen Su and Dhananjay Thiruvady and Andy Song and Frank Neumann and Carla Silva",
• pages = "914--923",
• address = "Melbourne, Australia",
• series = "GECCO '24",
• month = "14-18 " # jul,
• organisation = "SIGEVO",
• publisher = "Association for Computing Machinery",
• publisher_address = "New York, NY, USA",
• keywords = "genetic algorithms, genetic programming, linear GP, large language models, curriculum learning",
• isbn13 = "979-8-4007-0494-9",
• DOI = "doi:10.1145/3638529.3654056",
• size = "10 pages",
• abstract = "Genetic programming (GP) is a popular problem-solving and optimization technique. However, generating effective test cases for training and evaluating GP programs requires strong domain knowledge. Furthermore, GP programs often prematurely converge on local optima when given excessively difficult problems early in their training. Curriculum learning (CL) has been effective in addressing similar issues across different reinforcement learning (RL) domains, but it requires the manual generation of progressively difficult test cases as well as their careful scheduling. In this work, we leverage the domain knowledge and the strong generative abilities of large language models (LLMs) to generate effective test cases of increasing difficulties and schedule them according to various curricula. We show that by integrating a curriculum scheduler with LLM-generated test cases we can effectively train a GP agent player with environments-based curricula for a single-player game and opponent-based curricula for a multi-player game. Finally, we discuss the benefits and challenges of implementing this method for other problem domains.",
• notes = "GECCO-2024 GP A Recombination of the 33rd International Conference on Genetic Algorithms (ICGA) and the 29th Annual Genetic Programming Conference (GP)",

}

Genetic Programming entries for Steven P Jorgensen Giorgia Nadizar Gloria Pietropolli Luca Manzoni Eric Medvet Una-May O'Reilly Erik Hemberg

Citations