Efficient Calibration of Multi-Agent Simulation Models from Output Series with Bayesian Optimization

TL;DR

This paper introduces an efficient Bayesian optimization framework for calibrating multi-agent simulation models based solely on output series, addressing non-identifiability and high-dimensional distribution comparison issues.

Contribution

The paper proposes a novel calibration framework using eligibility sets, a generalized K-S test for high-dimensional data, and Bayesian optimization techniques.

Findings

Effective calibration on a multi-agent financial market simulator

Outperforms traditional methods in efficiency and accuracy

Addresses non-identifiability in simulation parameters

Abstract

Multi-agent simulation is commonly used across multiple disciplines, specifically in artificial intelligence in recent years, which creates an environment for downstream machine learning or reinforcement learning tasks. In many practical scenarios, however, only the output series that result from the interactions of simulation agents are observable. Therefore, simulators need to be calibrated so that the simulated output series resemble historical -- which amounts to solving a complex simulation optimization problem. In this paper, we propose a simple and efficient framework for calibrating simulator parameters from historical output series observations. First, we consider a novel concept of eligibility set to bypass the potential non-identifiability issue. Second, we generalize the two-sample Kolmogorov-Smirnov (K-S) test with Bonferroni correction to test the similarity between two high-dimensional distributions, which gives a simple yet effective distance metric between the output series sample sets. Third, we suggest using Bayesian optimization (BO) and trust-region BO (TuRBO) to minimize the aforementioned distance metric. Finally, we demonstrate the efficiency of our framework using numerical experiments both on a multi-agent financial market simulator.