The rapid development of machine learning has significantly influenced the field of evolutionary optimization. Traditional evolutionary optimization approaches often suffer from cold starts and inefficient solution searches. Learning-based evolutionary optimization, combining machine learning and evolutionary algorithms, has emerged as a promising research area (e.g., learning for EC appeared for the first time as a new track at GECCO2024). These methods enable the discovery of effective feature representations, prediction of near-optimal solutions, dimensionality reduction, and extraction of knowledge from training data to guide the search process. Despite these advancements, learning-based methods typically require large amounts of training data and often lack generalizability across different problem domains. The recent rise of large language models (LLMs) introduces a new paradigm, as these models are pre-trained on vast amounts of data and can be fine-tuned with a minimal number of examples to generate effective heuristic algorithms and solutions for evolutionary optimization.

This tutorial will provide an overview of the latest research development in intelligent evolutionary optimization, focusing on the transition from traditional machine learning to deep learning, and then to business optimization and evolutionary optimization guided by large language models.