📄 中文摘要
决策者利用半导体激光器产生的光子混沌动态，为解决多臂赌博机问题提供了一种超快速的方法，使用时间光信号作为顺序决策的驱动源。在这种系统中，混沌波形的采样间隔影响生成时间序列的时间相关性，实验表明决策准确性与自相关特性密切相关。然而，自相关的益处是否可以通过一个最简数学模型来解释仍然不清楚。研究分析了一种基于随机过程的时间序列决策模型，采用拉锯战原则解决双臂赌博机问题，揭示了自相关在决策过程中的重要性。
随机过程模型中自相关效应在基于时间序列的决策中的应用
📄 English Summary
Autocorrelation effects in a stochastic-process model for decision making via time series
Decision makers leveraging photonic chaotic dynamics generated by semiconductor lasers offer an ultrafast method for addressing multi-armed bandit problems, utilizing a temporal optical signal as the driving source for sequential decisions. In such systems, the sampling interval of the chaotic waveform shapes the temporal correlation of the resulting time series, with experiments indicating that decision accuracy is strongly dependent on this autocorrelation property. However, it remains uncertain whether the advantages of autocorrelation can be explained by a minimal mathematical model. This study analyzes a stochastic-process model of time-series-based decision making using the tug-of-war principle to solve the two-armed bandit problem, highlighting the significance of autocorrelation in the decision-making process.