What is the "standard score" regularizer?

[Hanxi1208]

I am working on reproducing the experiment described in the supplementary materials, specifically in Section S2.1.4, titled "Disentanglement of aleatoric and epistemic uncertainty" (Figure S3). In that section, the paper said used a "standard score" to replace the L1 error in the regularization term. However, the exact formula for this standard score is not explicitly provided in the paper.

Is the standard score you used for the regularizer defined as follows? $\frac{|y_i - \gamma|}{\sqrt{\frac{\beta}{\alpha-1} + \frac{\beta}{\nu(\alpha-1)}}}$
So the regularizer term is $L_i^R = \frac{|y_i - \gamma|}{\sqrt{\frac{\beta}{\alpha-1} + \frac{\beta}{\nu(\alpha-1)}}} * (2\nu + \alpha)$