Update continuous and discrete unsupervised learning slides

camlab-bioml · May 16, 2024 · ba64525 · ba64525
1 parent 5b593c2
commit ba64525
Show file tree

Hide file tree

Showing 2 changed files with 76 additions and 33 deletions.
diff --git a/slides/22-unsupervised_learning_continuous.html b/slides/22-unsupervised_learning_continuous.html
@@ -4,7 +4,7 @@
     <title>Unsupervised learning: continuous latent variable models</title>
     <meta charset="utf-8" />
     <meta name="author" content="Kieran Campbell" />
-    <script src="libs/header-attrs-2.11/header-attrs.js"></script>
+    <script src="libs/header-attrs-2.14/header-attrs.js"></script>
     <link href="libs/remark-css-0.0.1/default.css" rel="stylesheet" />
     <link href="libs/remark-css-0.0.1/metropolis.css" rel="stylesheet" />
     <link href="libs/remark-css-0.0.1/tamu-fonts.css" rel="stylesheet" />
@@ -13,11 +13,21 @@
     <textarea id="source">
 class: center, middle, inverse, title-slide
 
+.title[
 # Unsupervised learning: continuous latent variable models
+]
+.subtitle[
 ## Foundational Computational Biology II
+]
+.author[
 ### Kieran Campbell
+]
+.institute[
 ### Lunenfeld Tanenbaum Research Institute &amp; University of Toronto
-### 2021-02-22 (updated: 2022-04-04)
+]
+.date[
+### 2021-02-22 (updated: 2024-05-16)
+]
 
 ---
 
@@ -31,7 +41,7 @@
 # What we'll cover
 
 
-1. A historical perspective on continuous latent variable models
+1. Intuitive perspective on continuous latent variable models
 2. Principal component analysis and pPCA
 3. Non-negative matrix factorization
 4. Nonlinear methods: tSNE &amp; autoencoders
@@ -83,11 +93,11 @@
 
 ---
 
-# Origins of continuous latent variable models
+# Intuitive approach to latent variable models
 
-Spearman [Spe04] noted in 1904 correlation between pupil's school grades:
+What to quantify correlations between personality characteristics: 
 
-| Student |  Maths |  English |  Science | 
+| Employee |  Persistence |  Systematic planning |  Goal setting | 
 | -- | -- | -- | -- |
 | Alice | 68 | 79 | 65 |
 | Bob | 56 | 47 | 60 | 
@@ -96,7 +106,7 @@
 
 --
 
-&gt; What if the grades aren't just correlated with each other, but depend on an unseen ("latent") variable...called _general intelligence_ ?
+&gt; What if the characteristics aren't just correlated with each other, but depend on an unseen ("latent") variable (we might call this performance?)
 
 
 
@@ -108,9 +118,9 @@
 
 --
 
-* `\(y_{np}\)` grade of pupil `\(n\)` in subject `\(p\)`
-* `\(z_n\)` "intelligence" pupil `\(n\)`
-* `\(w_{p}\)` describes how the intelligence impacts the grade
+* `\(y_{np}\)` score of employee `\(n\)` in characteristic `\(p\)`
+* `\(z_n\)` "performance" employee `\(n\)`
+* `\(w_{p}\)` describes how the characteristic impacts the performance
 
 --
 
@@ -130,7 +140,7 @@
 
 # Modern applications
 
-Where do such models appear today?
+Where do such models appear in biology?
 
 --
 
@@ -515,11 +525,11 @@
 
 --
 
-Think of all _trinucleotide_ substitutions, e.g. `ACG` &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `AAG` 
+Think of all _trinucleotide_ substitutions, e.g. `ACG` &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `AAG` 
 
 --
 
-There are `\(96\)` possible mutations &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(N \times 96\)` count matrix
+There are `\(96\)` possible mutations &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(N \times 96\)` count matrix
 
 --
 
@@ -569,8 +579,8 @@
 
 So far all mappings have been linear:
 
-* `\(z\)` &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(y\)` via `\(\mathbf{y} = \mathbf{W}^\mathrm{T} \mathbf{z}\)` (e.g. pPCA, NMF)
-* `\(y\)` &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(z\)` via `\(z = \mathbf{u}^\mathrm{T} \mathbf{y}\)` (PCA)
+* `\(z\)` &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(y\)` via `\(\mathbf{y} = \mathbf{W}^\mathrm{T} \mathbf{z}\)` (e.g. pPCA, NMF)
+* `\(y\)` &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; `\(z\)` via `\(z = \mathbf{u}^\mathrm{T} \mathbf{y}\)` (PCA)
 
 but we may wish to learn more flexible, nonlinear embeddings that capture more complex structure
 
@@ -718,16 +728,13 @@
 non-negative factor model with optimal utilization of error estimates
 of data values". In: _Environmetrics_ 5.2, pp. 111-126.
 
-Spearman, C. (1904). "" General Intelligence" Objectively Determined
-and Measured."
-
 Van der Maaten, L. and G. Hinton (2008). "Visualizing data using
 t-SNE." In: _Journal of machine learning research_ 9.11.
 
 
     </textarea>
 <style data-target="print-only">@media screen {.remark-slide-container{display:block;}.remark-slide-scaler{box-shadow:none;}}</style>
-<script src="https://remarkjs.com/downloads/remark-latest.min.js"></script>
+<script src="libs/remark-latest.min.js"></script>
 <script>var slideshow = remark.create({
 "highlightStyle": "github",
 "highlightLines": true,
@@ -772,6 +779,19 @@
     deleted = true;
   });
 })();
+// add `data-at-shortcutkeys` attribute to <body> to resolve conflicts with JAWS
+// screen reader (see PR #262)
+(function(d) {
+  let res = {};
+  d.querySelectorAll('.remark-help-content table tr').forEach(tr => {
+    const t = tr.querySelector('td:nth-child(2)').innerText;
+    tr.querySelectorAll('td:first-child .key').forEach(key => {
+      const k = key.innerText;
+      if (/^[a-z]$/.test(k)) res[k] = t;  // must be a single letter (key)
+    });
+  });
+  d.body.setAttribute('data-at-shortcutkeys', JSON.stringify(res));
+})(document);
 (function() {
   "use strict"
   // Replace <script> tags in slides area to make them executable

diff --git a/slides/22-unsupervised_learning_discrete.html b/slides/22-unsupervised_learning_discrete.html
@@ -4,7 +4,7 @@
     <title>Unsupervised learning: mixture models</title>
     <meta charset="utf-8" />
     <meta name="author" content="Kieran Campbell" />
-    <script src="libs/header-attrs-2.11/header-attrs.js"></script>
+    <script src="libs/header-attrs-2.14/header-attrs.js"></script>
     <link href="libs/remark-css-0.0.1/default.css" rel="stylesheet" />
     <link href="libs/remark-css-0.0.1/metropolis.css" rel="stylesheet" />
     <link href="libs/remark-css-0.0.1/tamu-fonts.css" rel="stylesheet" />
@@ -13,11 +13,21 @@
     <textarea id="source">
 class: center, middle, inverse, title-slide
 
+.title[
 # Unsupervised learning: mixture models
+]
+.subtitle[
 ## Foundational Computational Biology II
+]
+.author[
 ### Kieran Campbell
+]
+.institute[
 ### Lunenfeld Tanenbaum Research Institute &amp; University of Toronto
-### 2021-02-22 (updated: 2022-04-04)
+]
+.date[
+### 2021-02-22 (updated: 2024-05-16)
+]
 
 ---
 
@@ -142,13 +152,13 @@
 
 `\(p(\texttt{Data} | z_n=c)\)` 
 
-  &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _what is the distribution of the data given it's assigned to cluster `\(c\)`?_
+  &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _what is the distribution of the data given it's assigned to cluster `\(c\)`?_
 
 --
 
 `\(p(z_n=c) = \pi_c\)` 
 
-  &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _what's my prior expectation sample `\(n\)` originates from cluster `\(c\)`?_
+  &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _what's my prior expectation sample `\(n\)` originates from cluster `\(c\)`?_
 
 ---
 
@@ -201,7 +211,7 @@
 ```
 
 ```
-## [1] 1 1 2 2 2 2
+## [1] 1 1 1 2 1 1
 ```
 
 ---
@@ -375,7 +385,7 @@
 ```
 
 ```
-## number of iterations= 6
+## number of iterations= 8
 ```
 
 ---
@@ -392,7 +402,7 @@
 ```
 
 ```
-## [1] -0.9647411  1.9973481
+## [1] -0.967592  1.998353
 ```
 
 * `\(\sigma_1 = 0.5\)`, `\(\sigma_2 = 0.1\)`
@@ -403,7 +413,7 @@
 ```
 
 ```
-## [1] 0.4891417 0.1000087
+## [1] 0.50871571 0.09321453
 ```
 
 * `\(p(z_n = 1) = \pi_1 = 1 - \pi_2 = 0.7\)`
@@ -414,7 +424,7 @@
 ```
 
 ```
-## [1] 0.728 0.272
+## [1] 0.7090005 0.2909995
 ```
 
 
@@ -427,17 +437,17 @@
 
 --
 
-In our example: 1 &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 2 and 2 &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 1
+In our example: 1 &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 2 and 2 &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 1
 
 --
 
 Requires care in analysis, e.g. imagine we're clustering bulk tumour gene expression
 
-* Cluster 1 &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _hypoxic_ cluster
+* Cluster 1 &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _hypoxic_ cluster
 
-* Cluster 2 &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _proliferative_ cluster
+* Cluster 2 &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; _proliferative_ cluster
 
-but you rerun the clustering and &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 2 and 2 &lt;svg viewBox="0 0 448 512" xmlns="http://www.w3.org/2000/svg" style="height:1em;fill:currentColor;position:relative;display:inline-block;top:.1em;"&gt;  [ comment ]  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 1
+but you rerun the clustering and &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 2 and 2 &lt;svg viewBox="0 0 448 512" style="height:1em;position:relative;display:inline-block;top:.1em;" xmlns="http://www.w3.org/2000/svg"&gt;  &lt;path d="M190.5 66.9l22.2-22.2c9.4-9.4 24.6-9.4 33.9 0L441 239c9.4 9.4 9.4 24.6 0 33.9L246.6 467.3c-9.4 9.4-24.6 9.4-33.9 0l-22.2-22.2c-9.5-9.5-9.3-25 .4-34.3L311.4 296H24c-13.3 0-24-10.7-24-24v-32c0-13.3 10.7-24 24-24h287.4L190.9 101.2c-9.8-9.3-10-24.8-.4-34.3z"&gt;&lt;/path&gt;&lt;/svg&gt; 1
 
 --
 
@@ -593,7 +603,7 @@
 
     </textarea>
 <style data-target="print-only">@media screen {.remark-slide-container{display:block;}.remark-slide-scaler{box-shadow:none;}}</style>
-<script src="https://remarkjs.com/downloads/remark-latest.min.js"></script>
+<script src="libs/remark-latest.min.js"></script>
 <script>var slideshow = remark.create({
 "highlightStyle": "github",
 "highlightLines": true,
@@ -638,6 +648,19 @@
     deleted = true;
   });
 })();
+// add `data-at-shortcutkeys` attribute to <body> to resolve conflicts with JAWS
+// screen reader (see PR #262)
+(function(d) {
+  let res = {};
+  d.querySelectorAll('.remark-help-content table tr').forEach(tr => {
+    const t = tr.querySelector('td:nth-child(2)').innerText;
+    tr.querySelectorAll('td:first-child .key').forEach(key => {
+      const k = key.innerText;
+      if (/^[a-z]$/.test(k)) res[k] = t;  // must be a single letter (key)
+    });
+  });
+  d.body.setAttribute('data-at-shortcutkeys', JSON.stringify(res));
+})(document);
 (function() {
   "use strict"
   // Replace <script> tags in slides area to make them executable