Geometric Semantic Genetic Programming (GSGP) is a variant of Genetic Programming (GP) that induces an error surface without local minima for supervised learning tasks. However, GSGP is limited by the fact that its operators produce increasingly large individuals, leading to overly complex models. The slim addresses this issue by introducing a deflate geometric semantic mutation capable of producing offspring smaller than their parents. Preliminary studies show that slim can maintain accuracy comparable to traditional GSGP while generating much smaller models. However, a thorough analysis of this mutation remains lacking. This work fills that gap by conducting a detailed study of the deflate mutation, focusing on its behavior and practical value. Our results show that, when applied at the right stage of evolution, deflate mutation mitigates overfitting and yields compact, accurate models. This is also the first study to explore the timing and interaction of inflate and deflate mutations in slim, demonstrating how deflation enhances generalization and reduces overfitting. We support our conclusions with a comprehensive experimental approach, including comparisons between exclusive use of inflate mutation and alternating it with deflation. We also evaluate numerical indicators such as improvement rate and training effectiveness. The consistency across these methods reinforces our findings and highlights the deflate mutation as a robust regularization strategy. Finally, when compared to established non-evolutionary machine learning methods, SLIM shows competitive performance. Overall, this study confirms SLIM as a promising direction for GP and lays the foundation for future research.

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.

This paper presents slim_gsgp: an open-source Python library that provides the first ever framework for the Semantic Learning algorithm based on Inflate and deflate Mutation (SLIM-GSGP). Proposed in 2024, SLIM-GSGP is a promising non-bloating variant of Geometric Semantic Genetic Programming (GSGP). slim_gsgp includes all existing SLIM-GSGP variants, as well as traditional GSGP and standard Genetic Programming (GP), facilitating comparative analysis and benchmarking. Additionally, slim_gsgp's parallel computation and semi-modular architecture renders it not only fast but also user-friendly and easily extensible, thereby serving as a valuable resource for researchers aiming to advance this emerging and promising area of research. The source code and documentation can be accessed at https://github.com/DALabNOVA/slim.

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.

Cellular automata (CA) are a convenient way to describe the distributed evolution of a dynamical system over discrete time and space.They can be used to express evolutionary algorithms (EAs), where the time is the flow of iterations and the space is where the population is hosted.When the CA evolves over a finite grid of cells, the substrate, each cell hosts an individual and the CA rule applies variation operators using the local and neighbor individuals.In this paper, we explore the possibility of enforcing a structure on the substrate.Instead of a flat toroidal grid, we use substrates where some empty cells never host individuals.These cells may act as barriers, slowing down the propagation of genetic traits and hence potentially improving the population diversity, eventually mitigating the risk of premature convergence.We experimentally evaluate the impact of these substrates using a simple CA-based EA on multi-modal and multi-objective problems.We find evidence of a positive impact in some circumstances; on multi-modal problems, convergence is slightly faster and the EA more often reaches all the targets.