Algorithm 4: The ABS ratio $\alpha$ and the CRE bias $\beta$ are given to optimize the number of SBS activation $\left|S\right|$.

Require: $a_t=(\alpha ,\beta ,\left|S\right|)$.

Ensure: Optimized ABS ratio $\alpha$.

1: Initialize $Q_{\left|S\right|}(s,a)$, state s and n = 0;

2: Setting learning rate ${\lambda }_3$, greedy probability ${\epsilon }_3$, discount factor ${\gamma }_3$, and $threshol{d}_3$

3: while n <= $threshol{d}_3$ do

4:  In state s, select the optimal action a with greedy probability ${\epsilon }_3$;

5:  Observe r;

6:  randomly transfer from s to $s^{{}^{\prime }}$;

7:  Update $Q_{\left|S\right|}(s,a)$ according to Formula (20);

8:  $s\leftarrow s^{{}^{\prime }}$;

9:  $n=n+1$;

10: end while

11: Output: $\left|S\right|=a$;