The Study of Reinforcement Learning for Traffic Self-Adaptive Control under Multiagent Markov Game Environment

Figure 3

𝑄-values of TSCA 1, TSCA 2, and TSCA 5 per minute under 𝑠 and in 𝑎.