$\alpha = \frac{1}{1 + decay\_rate \cdot epoch\_num} \alpha_0$, where $\alpha$ is the learning rate in the current epoch, $\alpha_0$ is the initial learning rate, $epoch\_num$ is the current epoch, and $decay\_rate$ is the selected decay rate. The decay rate is a tunable hyperparameter. Initializing $decay\_rate = 1$ and $\alpha_0 = 0.2$, we can graph an example with $epoch\_num$ on the x-axis and $\alpha$ on the y-axis to observe the decay of the learning rate.