diff --git a/chapter-prelude/prelude.ipynb b/chapter-prelude/prelude.ipynb
index 5ab4dcbc..65c7d827 100644
--- a/chapter-prelude/prelude.ipynb
+++ b/chapter-prelude/prelude.ipynb
@@ -51,7 +51,7 @@
     "\n",
     "- **Part 2: Modelling approximation uncertainty: Set-based versus distributional representations.** Different statistical approaches for modelling uncertainty are discussed in {ref}`Chapter 4 <sbvd>`.\n",
     "\n",
-    "- **Part 3: Machine learning methods for representing uncertainty.** This is the main part of the book, which presents several different machine learning methods that allow for representing a learners uncertainty in a prediction. First, approaches that use classical frequentist statistics for quantifying uncertainty are discussed: Chapters {ref}`5<pe-scoring>`, {ref}`6<pe-calibration>` and {ref}`7<pe-ensemble>` discuss how to estimate probabilities via scoring, calibration and ensembles. {ref}`Chapter 8<fisher>` treat maximum likelihood estimation and the fisher information matrix. {ref}`Chapter 9<genmodels>` discusses generative models. Next, Bayesian approaches for uncertainty quantification are discussed: {ref}`Chapter 10<gaussian-processes>` presents gaussian processes. Chapter {ref}`11<deep-ensembles>` and {ref}`12<bayesian-nn>` describe ensembles of deep neural networks and bayesian neural networks. {ref}`Chapter 13<credal>` addresses the concept of credal sets and {ref}`chapter 14<uqnl>` reliable classification. Lastly, the concept of set valued prediction is introduced. Chapter {ref}`15<conformal-class>` and {ref}`16<conformal-reg>` discuss conformal prediction for classification and regression respectively. {ref}`Chapter 17<set-util>` explains set-valued prediction based on utility maximization.\n",
+    "- **Part 3: Machine learning methods for representing uncertainty.** This is the main part of the book, which presents several different machine learning methods that allow for representing a learners uncertainty in a prediction. First, approaches that use classical frequentist statistics for quantifying uncertainty are discussed: Chapters {ref}`5<pe-scoring>`, {ref}`6<pe-calibration>` and {ref}`7<pe-ensemble>` discuss how to estimate probabilities via scoring, calibration and ensembles. {ref}`Chapter 8<fisher>` treat maximum likelihood estimation and the fisher information matrix. {ref}`Chapter 9<genmodels>` discusses generative models. Next, Bayesian approaches for uncertainty quantification are discussed: {ref}`Chapter 10<gaussian-processes>` presents gaussian processes. Chapter {ref}`11<deep-ensembles>` and {ref}`12<bayesian-nn>` describe ensembles of deep neural networks and bayesian neural networks. {ref}`Chapter 13<credal>` addresses the concept of credal sets and {ref}`chapter 14<uqnl>` reliable classification. Lastly, the concept of set valued prediction is introduced. Chapter {ref}`15<conformal-class>` and [16](../chapter-conformel_regression/conformel_regression) discuss conformal prediction for classification and regression respectively. {ref}`Chapter 17<set-util>` explains set-valued prediction based on utility maximization.\n",
     "\n",
     "</div>"
    ]