Update 6-uncertainty-overview.md

episodes/6-uncertainty-overview.md

@@ -23,8 +23,7 @@ exercises: 0
 - Summarize when and where different uncertainty estimation methods are most useful.  
 
 ::::::::::::::::::::::::::::::::::::::::::::::::
-
-## How confident is my model? Will it generalize to new data or subpopulations?
+### How confident is my model? Will it generalize to new data or subpopulations?
 Understanding how confident a model is in its predictions is a valuable tool for building trustworthy AI systems, especially in high-stakes settings like healthcare or autonomous vehicles. Model uncertainty estimation focuses on quantifying the model's confidence and is often used to identify predictions that require further review or caution.
 
 ### Sources of uncertainty 
@@ -162,4 +161,4 @@ When choosing a method, it’s important to consider the trade-offs in computati
 3. **Out-of-distribution detection**: Identify inputs outside the training distribution.  
    - Example application: Flagging out-of-scope queries in chatbot systems.  
    - Reference: Hendrycks, D., & Gimpel, K. (2017). "A baseline for detecting misclassified and out-of-distribution examples in neural networks."  
-     [ArXiv](https://arxiv.org/abs/1610.02136).
+     [ArXiv](https://arxiv.org/abs/1610.02136).