Formal  Model and Analysis for the Random Event in the Intelligent Car with Stochastic Petri Nets and Z

Table 1

Advantages and disadvantages of three reinforcement learning algorithms.

Category	Representative algorithm	Advantage	Disadvantage
Value learning	DQN	It can solve high-dimensional complex problems and is not easy to fall into local optimum	Overfitting, low sample utilization, instability, poor convergence
Policy learning	Monte Carlo method	High stability and strong convergence	Large variance, slow convergence, the easy local optimal solution
Value policy learning	Actor-critic	Small variance, fast training, and can solve continuous problems	Low learning efficiency, large deviation, and poor stability