diff --git a/README.md b/README.md
index b6649f6..6fc941b 100644
--- a/README.md
+++ b/README.md
@@ -156,7 +156,7 @@ C. Oballe and V. Maroulas provide a
 [tutorial](https://github.com/coballejr/misc/blob/master/Tuto-GUDHI-bayes-tda.ipynb)
 for a Python module that implements the model for Bayesian inference
 with persistence diagrams introduced in their
-[paper](https://epubs.siam.org/doi/pdf/10.1137/19M1268719).
+[paper](https://epubs.siam.org/doi/abs/10.1137/19M1268719).
 
 ### 06 - Machine learning and deep learning with TDA
 
diff --git a/Tuto-GUDHI-persistence-diagrams.ipynb b/Tuto-GUDHI-persistence-diagrams.ipynb
index 2bffd32..237b822 100644
--- a/Tuto-GUDHI-persistence-diagrams.ipynb
+++ b/Tuto-GUDHI-persistence-diagrams.ipynb
@@ -897,7 +897,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We apply a Multidimensional Scaling method (from the [<code>scikit-learn</code>](http://scikit-learn.org/stable/) library)  to visualize the Bottleneck distances:"
+    "We apply a Multidimensional Scaling method (from the [<code>scikit-learn</code>](https://scikit-learn.org/stable/) library)  to visualize the Bottleneck distances:"
    ]
   },
   {
diff --git a/Tuto-GUDHI-persistent-entropy.ipynb b/Tuto-GUDHI-persistent-entropy.ipynb
index d29fb49..50c4c1c 100644
--- a/Tuto-GUDHI-persistent-entropy.ipynb
+++ b/Tuto-GUDHI-persistent-entropy.ipynb
@@ -340,7 +340,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Once all bars in our barcodes are finite, we can apply PE. We use the [representations.Entropy](http://gudhi.gforge.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.Entropy) method. When applied to a list of barcodes, it gives another list with their respective persistent entropys."
+    "Once all bars in our barcodes are finite, we can apply PE. We use the [representations.Entropy](https://gudhi.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.Entropy) method. When applied to a list of barcodes, it gives another list with their respective persistent entropys."
    ]
   },
   {
@@ -622,7 +622,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We can calculate the Betti curves of these barcodes using the [representations.BettiCurve](http://gudhi.gforge.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.BettiCurve) method. When applied to a list of barcodes, it gives another list with their respective Betti curves."
+    "We can calculate the Betti curves of these barcodes using the [representations.BettiCurve](https://gudhi.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.BettiCurve) method. When applied to a list of barcodes, it gives another list with their respective Betti curves."
    ]
   },
   {
@@ -681,7 +681,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We proceed similarly to calculate the ES function. We use the [representations.Entropy](http://gudhi.gforge.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.Entropy) method. This time, we include the parameter mode='vector' to specify we do not want to calculate PE but ES. Using a list of barcodes as an input, we obtain a list of curves, each of which being the corresponding ES function. You can use the parameter sample_range to specify the domain of the function and resolution to specify the number of samples you use to calculate it. The normalized parameter indicates if you want to normalize the ES function by its L1 norm (it is done by default). It is used when you want your function to be independent of the coordinates used to generate the persistence barcode."
+    "We proceed similarly to calculate the ES function. We use the [representations.Entropy](https://gudhi.inria.fr/python/latest/representations.html#gudhi.representations.vector_methods.Entropy) method. This time, we include the parameter mode='vector' to specify we do not want to calculate PE but ES. Using a list of barcodes as an input, we obtain a list of curves, each of which being the corresponding ES function. You can use the parameter sample_range to specify the domain of the function and resolution to specify the number of samples you use to calculate it. The normalized parameter indicates if you want to normalize the ES function by its L1 norm (it is done by default). It is used when you want your function to be independent of the coordinates used to generate the persistence barcode."
    ]
   },
   {
@@ -788,7 +788,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Recall you can use the ES function as a parameter in your scikit-learn pipeline, together with other methods from the [representations](https://gudhi.inria.fr/python/3.1.0.rc1/representations.html) module."
+    "Recall you can use the ES function as a parameter in your scikit-learn pipeline, together with other methods from the [representations](https://gudhi.inria.fr/python/latest/representations.html) module."
    ]
   }
  ],
diff --git a/Tuto-GUDHI-representations.ipynb b/Tuto-GUDHI-representations.ipynb
index f32f704..033317f 100644
--- a/Tuto-GUDHI-representations.ipynb
+++ b/Tuto-GUDHI-representations.ipynb
@@ -30,7 +30,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "In this notebook, we will see how to efficiently combine machine learning and topological data analysis with the Gudhi library and its [representations](https://gudhi.inria.fr/python/3.1.0.rc1/representations.html) module. We will see how to compute the various Hilbert representations of persistence diagrams and how to use them in order to classify a set of persistence diagrams! Ready? Let's go!"
+    "In this notebook, we will see how to efficiently combine machine learning and topological data analysis with the Gudhi library and its [representations](https://gudhi.inria.fr/python/latest/representations.html) module. We will see how to compute the various Hilbert representations of persistence diagrams and how to use them in order to classify a set of persistence diagrams! Ready? Let's go!"
    ]
   },
   {
@@ -329,7 +329,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Neat, right? Gudhi also contains implementations of less common vectorization methods, such as the [Betti curve](https://www.researchgate.net/publication/316604237_Time_Series_Classification_via_Topological_Data_Analysis), the [complex polynomial]( https://link.springer.com/chapter/10.1007%2F978-3-319-23231-7_27), or the [topological vector](https://diglib.eg.org/handle/10.1111/cgf12692). You can check the [representations](https://gudhi.inria.fr/python/3.1.0.rc1/representations.html) module to check everything that is available."
+    "Neat, right? Gudhi also contains implementations of less common vectorization methods, such as the [Betti curve](https://www.researchgate.net/publication/316604237_Time_Series_Classification_via_Topological_Data_Analysis), the [complex polynomial]( https://link.springer.com/chapter/10.1007%2F978-3-319-23231-7_27), or the [topological vector](https://diglib.eg.org/handle/10.1111/cgf12692). You can check the [representations](https://gudhi.inria.fr/python/latest/representations.html) module to check everything that is available."
    ]
   },
   {