Fixes FO00039-63: some minor documentation typos.

docs/source/safety-aspects/fairness.rst

@@ -39,7 +39,7 @@ rate (expectation of model predictions w.r.t. the sensitive feature), so that is
 model works equally well for various groups. It is stricter than demographic parity because it not only demands
 that the model's predictions are not influenced by sensitive group membership but also requires that the groups
 have the same rates of false positive and true positive predictions :cite:p:`fairness-Agarwal2018`.
-More formally, it equalized odds can be described by
+More formally, equalized odds can be described by
 
 .. math::
 

docs/source/safety-aspects/uncertainty.rst

@@ -11,7 +11,7 @@ Thetis analyzes and evaluates the quality and consistency of the uncertainty tha
 AI prediction in most cases (e.g., score or confidence information in classification, "objectness" score in object
 detection, or variance estimation in probabilistic regression settings).
 The evaluation of uncertainty quality depends on the selected task. We give a brief overview about the mathematical
-background and the used metrics in the following.
+background and the used metrics in the following section.
 
 Classification
 --------------
@@ -103,7 +103,7 @@ The uncertainty evaluation differs from standard classification evaluation in tw
    IoU describes to which degree predicted and existing objects need to overlap to be considered as matching. Thus,
    all evaluation results are given w.r.t. a certain IoU score.
 
-Furthermore, recent work have shown that the calibration error might also be position-dependent
+Furthermore, recent work has shown that the calibration error might also be position-dependent
 :cite:p:`uncertainty-Kueppers2020`, :cite:p:`uncertainty-Kueppers2022a`, i.e., the calibration properties of objects located in the center
 of an image might differ from objects located at the image boundaries.
 Thus, given an object detection model that estimates an object with label :math:`\hat{Y} \in \mathcal{Y}`,