diff --git a/slides-pdf/slides-mc-softmax-regression.pdf b/slides-pdf/slides-mc-softmax-regression.pdf
index 94871389..3924bd77 100644
Binary files a/slides-pdf/slides-mc-softmax-regression.pdf and b/slides-pdf/slides-mc-softmax-regression.pdf differ
diff --git a/slides/multiclass/slides-mc-softmax-regression.tex b/slides/multiclass/slides-mc-softmax-regression.tex
index 4649b5ac..fd6f485b 100644
--- a/slides/multiclass/slides-mc-softmax-regression.tex
+++ b/slides/multiclass/slides-mc-softmax-regression.tex
@@ -52,7 +52,7 @@
       f_k(\xv) = \thetav_k^\top \xv, \quad k = 1, 2, ..., g,
     $$
   each indicating the confidence in class $k$.
-  \item The $g$ score functions are transformed into $g$ probability functions by the \textbf{softmax} function $s:\R^g \to \R^g$ 
+  \item The $g$ score functions are transformed into $g$ probability functions by the \textbf{softmax} function $s:\R^g \to [0,1]^g$ 
 
   $$
     \pi_k(\xv) = s(\fx)_k = \frac{\exp(\thetav_k^\top \xv)}{\sum_{j = 1}^g \exp(\thetav_j^\top \xv) }\,,