Policy Search through Genetic Programming and LLM-assisted Curriculum Learning
Name
3772718.pdf
Size
1011.1 KB
Format
Adobe PDF
Checksum (MD5)
ee274a38789ce809930886ceadd37254
Author(s)
Jorgensen, Steven
Nadizar, Giorgia
Pietropolli, Gloria
Manzoni, Luca
Medvet, Eric
O'Reilly, Una-May
Hemberg, Erik
Date Issued
October 31, 2025
Journal
ACM Transactions on Evolutionary Learning and Optimization
Publisher
ACM
Citation
Steven Jorgensen, Giorgia Nadizar, Gloria Pietropolli, Luca Manzoni, Eric Medvet, Una-May O'Reilly, and Erik Hemberg. 2025. Policy Search through Genetic Programming and LLM-assisted Curriculum Learning. ACM Trans. Evol. Learn. Optim.
Version
Final published version
Abstract
Curriculum learning (CL) consists in using a diverse set of user-provided test cases, with varying levels of difficulty and organized in a suitable progression, for learning a policy. The quality of test cases is important to allow optimization techniques as genetic programming (GP) to solve policy search problems. In this work, we evaluate large language models (LLMs) as providers of test cases for GP-based policy search. We consider two policy search tasks, a single-player and a multi-player game, and four LLMs differing in complexity and specialization, which we prompt in order to generate suitable test cases for the two games. We experimentally assess the intrinsic quality of LLM-generated test cases and their utility when inserted in a curriculum consumed by a GP optimization. We evaluate the robustness of the approach with respect to the way cases are scheduled in curricula and with respect to the policy representation, for which we use both graphs and linear programs evolved by GP. We observe that the effectiveness of LLM-assisted CL depends on both the choice of LLM and the design of the prompting and scheduling strategies. These findings highlight important considerations for leveraging LLMs in automated curriculum design for GP-based optimization.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3772718
