Efficient Calibration of Multi-Agent Simulation Models from Output Series with Bayesian Optimization
This addresses the calibration challenge for researchers and practitioners using multi-agent simulations in fields like AI and finance, though it is an incremental improvement combining existing techniques.
The paper tackles the problem of calibrating multi-agent simulation models when only output series are observable, proposing a framework that uses a novel eligibility set, a generalized Kolmogorov-Smirnov test, and Bayesian optimization to efficiently match simulated outputs to historical data, demonstrating effectiveness in a financial market simulator.
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.