TY - GEN
T1 - A Quantitative Learning Method for Simulation Model Evaluation Using L-SHADE Optimized Structured Regression
AU - Zhang, Jiayi
AU - Laili, Yuanjun
AU - Gong, Jiabei
AU - Zhang, Lin
AU - Ren, Lei
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.
PY - 2026
Y1 - 2026
N2 - The evaluation of simulation models is a fundamental step in establishing their credibility. Conventional approaches often rely on qualitative expert judgment, a process that is often inefficient and labor-intensive. Additionally, simulation outputs are often limited due to high computational costs or intrinsic model properties. To address these challenges, we propose a Quantitative Learning Method (QLM) that employs a dual-layer architecture. This method is designed to integrate historical assessment scores in small-sample contexts, learning the mapping from raw simulation outputs to evaluation metrics. Validated on two classic simulation models against seven baseline algorithms, QLM demonstrates robust generalization and stable performance, mitigating the overfitting prevalent in machine learning approaches under data constraints. Furthermore, its inherent transparency the mapping between simulation outputs and credibility scores offers considerable interpretability and practical value.
AB - The evaluation of simulation models is a fundamental step in establishing their credibility. Conventional approaches often rely on qualitative expert judgment, a process that is often inefficient and labor-intensive. Additionally, simulation outputs are often limited due to high computational costs or intrinsic model properties. To address these challenges, we propose a Quantitative Learning Method (QLM) that employs a dual-layer architecture. This method is designed to integrate historical assessment scores in small-sample contexts, learning the mapping from raw simulation outputs to evaluation metrics. Validated on two classic simulation models against seven baseline algorithms, QLM demonstrates robust generalization and stable performance, mitigating the overfitting prevalent in machine learning approaches under data constraints. Furthermore, its inherent transparency the mapping between simulation outputs and credibility scores offers considerable interpretability and practical value.
KW - Differential Evolution algorithm
KW - Few-shot Learning
KW - Multi-task Regression
KW - Simulation Model Evaluation
UR - https://www.scopus.com/pages/publications/105023171611
U2 - 10.1007/978-981-95-4472-1_1
DO - 10.1007/978-981-95-4472-1_1
M3 - 会议稿件
AN - SCOPUS:105023171611
SN - 9789819544714
T3 - Communications in Computer and Information Science
SP - 3
EP - 15
BT - Methods and Applications for Modeling and Simulation of Complex Systems - 24th Asia Simulation Conference, AsiaSim 2025, Proceedings
A2 - Cai, Wentong
A2 - Low, Malcolm
A2 - Tan, Gary
A2 - D'Angelo, Gabriele
A2 - Ta, Duong
PB - Springer Science and Business Media Deutschland GmbH
T2 - 24th Asia Simulation Conference on Methods and Applications for Modeling and Simulation of Complex Systems, AsiaSim 2025
Y2 - 17 November 2025 through 19 November 2025
ER -