Evolutionary Generative Optimization: Towards Fully Data-Driven Evolutionary Optimization via Generative Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Tao, Sun, Kebin, Liang, Zhenyu, Cheng, Ran, Jin, Yaochu, Tan, Kay Chen
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918336255754240
author Jiang, Tao
Sun, Kebin
Liang, Zhenyu
Cheng, Ran
Jin, Yaochu
Tan, Kay Chen
author_facet Jiang, Tao
Sun, Kebin
Liang, Zhenyu
Cheng, Ran
Jin, Yaochu
Tan, Kay Chen
contents Recent advances in data-driven evolutionary algorithms (EAs) have demonstrated the potential of leveraging historical data to improve optimization accuracy and adaptability. Despite these advancements, existing methods remain reliant on handcrafted process-level operators. In contrast, Evolutionary Generative Optimization (EvoGO) is a fully data-driven framework designed from the objective level, enabling autonomous learning of the entire search process. EvoGO streamlines the evolutionary optimization process into three stages: data preparation, model training, and population generation. The data preparation stage constructs a pairwise dataset to enrich training diversity without incurring additional evaluation costs. During model training, a tailored generative model learns to transform inferior solutions into superior ones. In the population generation stage, EvoGO replaces traditional reproduction operators with a scalable and parallelizable generative mechanism. Extensive experiments on numerical benchmarks, classical control problems, and high-dimensional robotic tasks demonstrate that EvoGO consistently converges within merely 10 generations and substantially outperforms a wide spectrum of optimization approaches, including traditional EAs, Bayesian optimization, and reinforcement learning based methods. Code is available at: https://github.com/EMI-Group/evogo
format Preprint
id arxiv_https___arxiv_org_abs_2508_00380
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evolutionary Generative Optimization: Towards Fully Data-Driven Evolutionary Optimization via Generative Learning
Jiang, Tao
Sun, Kebin
Liang, Zhenyu
Cheng, Ran
Jin, Yaochu
Tan, Kay Chen
Neural and Evolutionary Computing
Recent advances in data-driven evolutionary algorithms (EAs) have demonstrated the potential of leveraging historical data to improve optimization accuracy and adaptability. Despite these advancements, existing methods remain reliant on handcrafted process-level operators. In contrast, Evolutionary Generative Optimization (EvoGO) is a fully data-driven framework designed from the objective level, enabling autonomous learning of the entire search process. EvoGO streamlines the evolutionary optimization process into three stages: data preparation, model training, and population generation. The data preparation stage constructs a pairwise dataset to enrich training diversity without incurring additional evaluation costs. During model training, a tailored generative model learns to transform inferior solutions into superior ones. In the population generation stage, EvoGO replaces traditional reproduction operators with a scalable and parallelizable generative mechanism. Extensive experiments on numerical benchmarks, classical control problems, and high-dimensional robotic tasks demonstrate that EvoGO consistently converges within merely 10 generations and substantially outperforms a wide spectrum of optimization approaches, including traditional EAs, Bayesian optimization, and reinforcement learning based methods. Code is available at: https://github.com/EMI-Group/evogo
title Evolutionary Generative Optimization: Towards Fully Data-Driven Evolutionary Optimization via Generative Learning
topic Neural and Evolutionary Computing
url https://arxiv.org/abs/2508.00380