index.xml

<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Sancar Adali - personal website on Sancar Adali - personal website</title>
    <link>https://adalisan.github.io/</link>
    <description>Recent content in Sancar Adali - personal website on Sancar Adali - personal website</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <copyright>&amp;copy; 2016 Sancar Adali</copyright>
    <lastBuildDate>Sun, 10 Mar 2019 00:00:00 +0000</lastBuildDate>
    <atom:link href="/" rel="self" type="application/rss+xml" />

    <item>
      <title>Seeded Graph Matching</title>
      <link>https://adalisan.github.io/publication/sgm/</link>
      <pubDate>Sun, 10 Mar 2019 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/sgm/</guid>
      <description></description>
    </item>

    <item>
      <title>graph matching and spaces</title>
      <link>https://adalisan.github.io/post/graph-matching-and-spaces/</link>
      <pubDate>Thu, 26 Jul 2018 21:06:09 -0400</pubDate>

      <guid>https://adalisan.github.io/post/graph-matching-and-spaces/</guid>
      <description>&lt;p&gt;Part of &lt;a href=&#34;{{&amp;lt; ref &amp;quot;publication/thesis.md&amp;quot; &amp;gt;}}&#34; target=&#34;_blank&#34;&gt;my dissertation research&lt;/a&gt; was on the graph matching problem. For solving this problem in real settings, it is useful to think of vertices as elements of a space with the edges determining some sort of locality, (I guess this can be formalized with topology ). Then the graph/vertex matching problem becomes sort of a point cloud matching problem where you don&amp;rsquo;t know where the points are  , just the neighborhood relationships between them ( several manifold learning uses these relationships to find the manifolds defined by the points ). That is why embedding the vertices of the two graph using dissimilarities between vertex pairs can be used to match graphs. This is how we have used &lt;a href=&#34;{{&amp;lt; ref &amp;quot;publication/SGMviaJOFC.md&amp;quot; &amp;gt;}}&#34; target=&#34;_blank&#34;&gt;JOFC&lt;/a&gt; for seeded graph matching.&lt;/p&gt;

&lt;p&gt;The same paradigm works with point cloud registration or matching manifolds. For example, see  &lt;a href=&#34;https://arxiv.org/pdf/1710.04087.pdf&#34; target=&#34;_blank&#34;&gt;this paper by Facebook researchers &lt;/a&gt; for matching between  monolingual word embeddings in different languages. They start by finding a rigid transformation  between the two word embeddings in different languages (say english and french), based on closest matches using Procrustes method. They then modify the distance between an english vector mapped french and a french vector to solve  the point density problem: hub words (that are common words in either corpora) will have emeddings that are close to many other words  will be matched to many other words , and easily confused . I think this also applies to  hub vertices that are connected to many other vertices in graph matching. The distance modification uses  the number of words in the neighborhood of a common word , so that more dense regions of the embedding spaces will see this cross-domain distance increase. Another possible solution to this nonhomogenous point density in the embedding spaces may be using hyperbolic embeddings which I will summarize in another post.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>ALADDIN</title>
      <link>https://adalisan.github.io/project/aladdin/</link>
      <pubDate>Mon, 15 May 2017 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/aladdin/</guid>
      <description>&lt;p&gt;Software Development &amp;amp; Research for interactive system that carries out training and search for events in big video collections. &lt;a href=&#34;https://www.iarpa.gov/index.php/research-programs/aladdin-video&#34; target=&#34;_blank&#34;&gt;IARPA website for the ALADDIN program&lt;/a&gt;&lt;/p&gt;
</description>
    </item>

    <item>
      <title>CVRG</title>
      <link>https://adalisan.github.io/project/cvrg/</link>
      <pubDate>Mon, 15 May 2017 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/cvrg/</guid>
      <description>&lt;p&gt;I was a researcher in the machine learning group for the Cardiovascular Research Grid project. We worked with the cardiologists who are custodians of a dataset involving patients with ICD implants. Our objective was to come up with robust, interpretable predictors that predicted either high risk of ICD utilization , or sudden cardiac death. An important challenge was to create predictors that both utilized both survival data and did not become biased due to not-missing-at-random data.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>Fidelity-Commensurability Tradeoff in Joint Embedding of Disparate Dissimilarities</title>
      <link>https://adalisan.github.io/publication/fidcommjofc/</link>
      <pubDate>Thu, 10 Nov 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/fidcommjofc/</guid>
      <description>&lt;p&gt;Let&amp;rsquo;s say our data does not consist of $n$ observations of $d$-dimensional vectors, but instead an $n \times n$ matrix, whose entries consist of the dissimilarities between the $n$ observations as measured by a bivariate function $\delta(.,.)$.  We refer to the computation of $n$ separate $d$-dimensional vectors/points  that is as consistent as possible with a dissimilarity matrices as &amp;ldquo;embedding&amp;rdquo; this dissimilarity matrix.&lt;/p&gt;

&lt;p&gt;Consider the embedding of $K$   dissimilarity(distance-like) matrices between $n$ objects (coming from $K$ different bivariate functions  $\delta_k(.,.)$ ) or observations of $n$  $K-tuples$ of similar objects. In any case the different dissimilarity matrices are related in a statistical or causal sense. We refer to this relation as &amp;ldquo;matched&amp;rdquo;. Say $K=2$. A joint  embedding of these dissimilarity matrices will result in $2n$ points that consist of $n$ matched points.&lt;/p&gt;

&lt;p&gt;To embed multiple dissimilarities  ${\Delta_k},k=1,2$  into a commensurate space, so that the information represented in different dissimality matrices can be used for a data exploitation task,  an omnibus dissimilarity matrix  $M \in \mathbb{R}^{nk \times nk}$  is constructed. Consider, for $K=2$,
$$
M=  \left[ \begin{array}{cc}
         \Delta_1 &amp;amp; L&lt;br /&gt;
        L^T  &amp;amp; \Delta_2
     \end{array}  \right]     \label{omnibus}
$$ where $L$ is an imputed or estimated matrix for the two dissimilarities between $n$ objects. The embedding of this matrix will result in $2n$ observations of $d$-dimensional vectors.&lt;/p&gt;

&lt;p&gt;If we minimize a common multidimensional scaling objective function known as raw stress to embed, the objective function can be decomposed into three terms that are interpretable.
Fidelity describes how well the embedding in commensurate space preserves the original dissimilarities ${\Delta_k},k=1,2$. These terms would be the only ones that are minimized   if we embedded the dissimilarity matrices separately.&lt;/p&gt;

&lt;p&gt;Commensurability is how well the embedding  in commensurate space preserves matchedness of matched observations. These terms correspond to the deviation from the real dissimilarities between matched observations and the distances between matched points in the embedding.&lt;/p&gt;

&lt;p&gt;Separability is how well the embedding  in commensurate space preserves the disparateness of unmatched observations.
These terms correspond to the deviation from the real dissimilarities between unmatched observations and the distances between unmatched points in the embedding.&lt;/p&gt;

&lt;p&gt;We show that a weight parameter that controls the terms&amp;rsquo; importance in the minimization of raw stress is essential for data exploitation task such as detection of new matched observations, given a set of  matched observations.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>Joint Optimization of Fidelity and Commensurability for Manifold Alignment and Graph Matching</title>
      <link>https://adalisan.github.io/publication/thesis/</link>
      <pubDate>Thu, 10 Nov 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/thesis/</guid>
      <description>&lt;p&gt;More detail can easily be written here using &lt;em&gt;Markdown&lt;/em&gt; and $\rm \LaTeX$ math code.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>Learning spatiotemporal features for infrared action recognition with 3d convolutional neural networks</title>
      <link>https://adalisan.github.io/publication/infracv/</link>
      <pubDate>Thu, 10 Nov 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/infracv/</guid>
      <description></description>
    </item>

    <item>
      <title>Seeded Graph Matching Via Joint Optimization of Fidelity and Commensurability</title>
      <link>https://adalisan.github.io/publication/sgmviajofc/</link>
      <pubDate>Thu, 10 Nov 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/sgmviajofc/</guid>
      <description></description>
    </item>

    <item>
      <title>Data Fusion from Disparate Dissimilarities and Joint Optimization of Fidelity and Commensurability</title>
      <link>https://adalisan.github.io/project/jofc/</link>
      <pubDate>Wed, 27 Apr 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/jofc/</guid>
      <description>&lt;p&gt;My PhD thesis was on joint embedding of disparate dissimilarities for solving various inference tasks such as transfer learning, match detection and graph matching. I have dealt with various issues such as robustness to noise and compared with alternative approaches such as Canonical Correlational Analysis.
&lt;a href=&#34;https://github.com/adalisan/JOFC-GraphMatch&#34; target=&#34;_blank&#34;&gt;Github repo&lt;/a&gt;&lt;/p&gt;
</description>
    </item>

    <item>
      <title>External Project</title>
      <link>https://adalisan.github.io/project/example-external-project/</link>
      <pubDate>Wed, 27 Apr 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/example-external-project/</guid>
      <description></description>
    </item>

    <item>
      <title>Seeded Graph Matching</title>
      <link>https://adalisan.github.io/project/sgm/</link>
      <pubDate>Wed, 27 Apr 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/sgm/</guid>
      <description>&lt;p&gt;This variant of the graph matching problem provides an opportunity for the application of JOFC and also the adaptation of a state-of-the-art graph matching algorithm.
My work on this variant of the graph matching problem constituted part of my dissertation research. This work is presented in &lt;a href=&#34;http://arxiv.org/abs/1209.0367&#34; target=&#34;_blank&#34;&gt;this&lt;/a&gt; and &lt;a href=&#34;http://arxiv.org/abs/1401.3813http://arxiv.org/abs/1401.3813&#34; target=&#34;_blank&#34;&gt;this&lt;/a&gt; arxiv papers.&lt;/p&gt;

&lt;p&gt;The code for the algorithm and analysis is at
&lt;a href=&#34;https://github.com/adalisan/SeededGraphMatch&#34; target=&#34;_blank&#34;&gt;https://github.com/adalisan/SeededGraphMatch&lt;/a&gt;&lt;/p&gt;
</description>
    </item>

    <item>
      <title>XDATA</title>
      <link>https://adalisan.github.io/project/xdata/</link>
      <pubDate>Wed, 27 Apr 2016 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/project/xdata/</guid>
      <description>

&lt;h4 id=&#34;darpa-program-for-organizing-big-data-analysis-and-visualization-efforts&#34;&gt;DARPA program for organizing big data analysis and visualization efforts.&lt;/h4&gt;

&lt;p&gt;The different data analysis tasks we accomplished are &lt;a href=&#34;http://www.cis.jhu.edu/~parky/XDATA/&#34; target=&#34;_blank&#34;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;See the &lt;a href=&#34;https://docs.google.com/presentation/d/1PjKrZE0zIsT6egd2p4jNx7XvaOxjYDkB_OP8tSZI6LY/pub?start=false&amp;amp;loop=false&amp;amp;delayms=5000&#34; target=&#34;_blank&#34;&gt;midpoint presentation&lt;/a&gt;
and &lt;a href=&#34;xdata_ppt1/XDATA-Midpoint.html&#34; target=&#34;_blank&#34;&gt;final presentation&lt;/a&gt; for the summer workshop.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>Migration</title>
      <link>https://adalisan.github.io/post/migrating-to-hugo/</link>
      <pubDate>Wed, 20 Apr 2016 10:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/post/migrating-to-hugo/</guid>
      <description>&lt;p&gt;I have decided to switch to the academic theme for hugo , as it makes much more sense for a researcher. Now I just need to collect all the things I have done over the years.&lt;/p&gt;
</description>
    </item>

    <item>
      <title>Writing content with Markdown, LaTeX, and Shortcodes</title>
      <link>https://adalisan.github.io/post/writing-markdown-latex/</link>
      <pubDate>Wed, 20 Apr 2016 10:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/post/writing-markdown-latex/</guid>
      <description>&lt;p&gt;Content can be written using &lt;a href=&#34;https://github.com/adam-p/markdown-here/wiki/Markdown-Cheatsheet&#34; target=&#34;_blank&#34;&gt;Markdown&lt;/a&gt;, &lt;a href=&#34;https://en.wikibooks.org/wiki/LaTeX/Mathematics&#34; target=&#34;_blank&#34;&gt;LaTeX math&lt;/a&gt;, and &lt;a href=&#34;http://gohugo.io/extras/shortcodes/&#34; target=&#34;_blank&#34;&gt;Hugo Shortcodes&lt;/a&gt;. Additionally, HTML may be used for advanced formatting.&lt;/p&gt;

&lt;p&gt;This article gives an overview of the most common formatting options.&lt;/p&gt;

&lt;h2 id=&#34;sub-headings&#34;&gt;Sub-headings&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;emphasis&#34;&gt;Emphasis&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;Italics with *asterisks* or _underscores_.

Bold with **asterisks** or __underscores__.

Combined emphasis with **asterisks and _underscores_**.

Strikethrough with ~~two tildes~~.
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;ordered-lists&#34;&gt;Ordered lists&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;1. First item
2. Another item
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;unordered-lists&#34;&gt;Unordered lists&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;* First item
* Another item
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;images&#34;&gt;Images&lt;/h2&gt;

&lt;p&gt;Images may be added to a page by placing them in your &lt;code&gt;static/img/&lt;/code&gt; folder and referencing them using one of the following two notations:&lt;/p&gt;

&lt;p&gt;A general image:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;![alternative text for search engines](/img/screenshot.png)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;A numbered figure with caption:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; figure src=&amp;quot;/img/screenshot.png&amp;quot; title=&amp;quot;Figure Caption&amp;quot; &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;links&#34;&gt;Links&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;[I&#39;m a link](https://www.google.com)
[A post]({{&amp;lt; ref &amp;quot;post/hi.md&amp;quot; &amp;gt;}})
[A publication]({{&amp;lt; ref &amp;quot;publication/hi.md&amp;quot; &amp;gt;}})
[A project]({{&amp;lt; ref &amp;quot;project/hi.md&amp;quot; &amp;gt;}})
[Another section]({{&amp;lt; relref &amp;quot;hi.md#who&amp;quot; &amp;gt;}})
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;emojis&#34;&gt;Emojis&lt;/h2&gt;

&lt;p&gt;See the &lt;a href=&#34;http://www.webpagefx.com/tools/emoji-cheat-sheet/&#34; target=&#34;_blank&#34;&gt;Emoji cheat sheet&lt;/a&gt; for available emoticons. The following serves as an example, but you should remove the spaces between each emoji name and pair of semicolons:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;I : heart : Academic : smile :
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;I ❤️ Academic 😄&lt;/p&gt;

&lt;h2 id=&#34;blockquote&#34;&gt;Blockquote&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;&amp;gt; This is a blockquote.
&lt;/code&gt;&lt;/pre&gt;

&lt;blockquote&gt;
&lt;p&gt;This is a blockquote.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;h2 id=&#34;footnotes&#34;&gt;Footnotes&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;I have more [^1] to say.

[^1]: Footnote example.
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;I have more &lt;sup class=&#34;footnote-ref&#34; id=&#34;fnref:1&#34;&gt;&lt;a href=&#34;#fn:1&#34;&gt;1&lt;/a&gt;&lt;/sup&gt; to say.&lt;/p&gt;

&lt;h2 id=&#34;code-highlighting&#34;&gt;Code highlighting&lt;/h2&gt;

&lt;p&gt;Pass the &lt;em&gt;language&lt;/em&gt; of the code, such as &lt;code&gt;python&lt;/code&gt;, as a parameter after three backticks:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;```python
# Example of code highlighting
input_string_var = input(&amp;quot;Enter some data: &amp;quot;)
print(&amp;quot;You entered: {}&amp;quot;.format(input_string_var))
```
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;Result:&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-python&#34;&gt;# Example of code highlighting
input_string_var = input(&amp;quot;Enter some data: &amp;quot;)
print(&amp;quot;You entered: {}&amp;quot;.format(input_string_var))
&lt;/code&gt;&lt;/pre&gt;

&lt;h3 id=&#34;highlighting-options&#34;&gt;Highlighting options&lt;/h3&gt;

&lt;p&gt;The Academic theme uses &lt;a href=&#34;https://highlightjs.org&#34; target=&#34;_blank&#34;&gt;highlight.js&lt;/a&gt; for source code highlighting, and highlighting is enabled by default for all pages. However, several configuration options are supported that allow finer-grained control over highlight.js.&lt;/p&gt;

&lt;p&gt;The following table lists the supported options for configuring highlight.js, along with their expected type and a short description. A &amp;ldquo;yes&amp;rdquo; in the &lt;strong&gt;config.toml&lt;/strong&gt; column means the value can be set globally in &lt;code&gt;config.toml&lt;/code&gt;, and a &amp;ldquo;yes&amp;rdquo; in the &lt;strong&gt;preamble&lt;/strong&gt; column means that the value can be set locally in a particular page&amp;rsquo;s preamble.&lt;/p&gt;

&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;option&lt;/th&gt;
&lt;th&gt;type&lt;/th&gt;
&lt;th&gt;description&lt;/th&gt;
&lt;th&gt;config.toml&lt;/th&gt;
&lt;th&gt;preamble&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;

&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;highlight&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;boolean&lt;/td&gt;
&lt;td&gt;enable/disable highlighting&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;&lt;code&gt;highlight_languages&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;slice&lt;/td&gt;
&lt;td&gt;choose additional languages&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;&lt;code&gt;highlight_style&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;string&lt;/td&gt;
&lt;td&gt;choose a highlighting style&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;td&gt;no&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;&lt;code&gt;highlight_version&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;string&lt;/td&gt;
&lt;td&gt;choose the highlight.js version&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;td&gt;no&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;

&lt;h4 id=&#34;option-highlight&#34;&gt;Option &lt;code&gt;highlight&lt;/code&gt;&lt;/h4&gt;

&lt;p&gt;The &lt;code&gt;highlight&lt;/code&gt; option allows enabling or disabling the inclusion of highlight.js, either globally or for a particular page. If the option is unset, it has the same effect as if you had specified &lt;code&gt;highlight = true&lt;/code&gt;. That is, the highlight.js javascript and css files will be included in every page. If you&amp;rsquo;d like to only include highlight.js files on pages that actually require source code highlighting, you can set &lt;code&gt;highlight = false&lt;/code&gt; in &lt;code&gt;config.toml&lt;/code&gt;, and then override it by setting &lt;code&gt;highlight = true&lt;/code&gt; in the preamble of any pages that require source code highlighting. Conversely, you could enable highlighting globally, and disable it locally for pages that do not require it. Here is a table that shows whether highlighting will be enabled for a page, based on the values of &lt;code&gt;highlight&lt;/code&gt; set in &lt;code&gt;config.toml&lt;/code&gt; and/or the page&amp;rsquo;s preamble.&lt;/p&gt;

&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;config.toml&lt;/th&gt;
&lt;th&gt;page preamble&lt;/th&gt;
&lt;th&gt;highlighting enabled for page?&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;

&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;unset or true&lt;/td&gt;
&lt;td&gt;unset or true&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;unset or true&lt;/td&gt;
&lt;td&gt;false&lt;/td&gt;
&lt;td&gt;no&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;false&lt;/td&gt;
&lt;td&gt;unset or false&lt;/td&gt;
&lt;td&gt;no&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;false&lt;/td&gt;
&lt;td&gt;true&lt;/td&gt;
&lt;td&gt;yes&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;

&lt;h4 id=&#34;option-highlight-languages&#34;&gt;Option &lt;code&gt;highlight_languages&lt;/code&gt;&lt;/h4&gt;

&lt;p&gt;The &lt;code&gt;highlight_languages&lt;/code&gt; option allows you to specify additional languages that are supported by highlight.js, but are not considered &amp;ldquo;common&amp;rdquo; and therefore are not supported by default. For example, if you want source code highlighting for Go and clojure in all pages, set &lt;code&gt;highlight_languages = [&amp;quot;go&amp;quot;, &amp;quot;clojure&amp;quot;]&lt;/code&gt; in &lt;code&gt;config.toml&lt;/code&gt;. If, on the other hand, you want to enable a language only for a specific page, you can set &lt;code&gt;highlight_languages&lt;/code&gt; in that page&amp;rsquo;s preamble.&lt;/p&gt;

&lt;p&gt;The &lt;code&gt;highlight_languages&lt;/code&gt; options specified in &lt;code&gt;config.toml&lt;/code&gt; and in a page&amp;rsquo;s preamble are additive. That is, if &lt;code&gt;config.toml&lt;/code&gt; contains, &lt;code&gt;highlight_languages = [&amp;quot;go&amp;quot;]&lt;/code&gt; and the page&amp;rsquo;s preamble contains &lt;code&gt;highlight_languages = [&amp;quot;ocaml&amp;quot;]&lt;/code&gt;, then javascript files for &lt;em&gt;both&lt;/em&gt; go and ocaml will be included for that page.&lt;/p&gt;

&lt;p&gt;If the &lt;code&gt;highlight_languages&lt;/code&gt; option is set, then the corresponding javascript files will be served from the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs server&lt;/a&gt;. To see a list of available languages, visit the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs page&lt;/a&gt; and search for links with the word &amp;ldquo;languages&amp;rdquo;.&lt;/p&gt;

&lt;p&gt;The &lt;code&gt;highlight_languages&lt;/code&gt; option provides an easy and convenient way to include support for additional languages to be severed from a CDN. If serving unmodified files from cdnjs doesn&amp;rsquo;t meet your needs, you can include javascript files for additional language support via one of the methods described in the getting started guide.&lt;/p&gt;

&lt;h4 id=&#34;option-highlight-style&#34;&gt;Option &lt;code&gt;highlight_style&lt;/code&gt;&lt;/h4&gt;

&lt;p&gt;The &lt;code&gt;highlight_style&lt;/code&gt; option allows you to select an alternate css style for highlighted code. For example, if you wanted to use the solarized-dark style, you could set &lt;code&gt;highlight_style = &amp;quot;solarized-dark&amp;quot;&lt;/code&gt; in &lt;code&gt;config.toml&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;If the &lt;code&gt;highlight_style&lt;/code&gt; option is unset, the default is to use the file &lt;code&gt;/css/highlight.min.css&lt;/code&gt;, either the one provided by the Academic theme, or else the one in your local &lt;code&gt;static&lt;/code&gt; directory.  The &lt;code&gt;/css/highlight.min.css&lt;/code&gt; file provided by Academic is equivalent to the &lt;code&gt;github&lt;/code&gt; style from highlight.js.&lt;/p&gt;

&lt;p&gt;If the &lt;code&gt;highlight_style&lt;/code&gt; option &lt;em&gt;is&lt;/em&gt; set, then &lt;code&gt;/css/highlight.min.css&lt;/code&gt; is ignored, and the corresponding css file will be served from the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs server&lt;/a&gt;. To see a list of available styles, visit the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs page&lt;/a&gt; and search for links with the word &amp;ldquo;styles&amp;rdquo;.&lt;/p&gt;

&lt;p&gt;See the &lt;a href=&#34;https://highlightjs.org/static/demo/&#34; target=&#34;_blank&#34;&gt;highlight.js demo page&lt;/a&gt; for examples of available styles.&lt;/p&gt;

&lt;p&gt;Missing alert
Not all styles listed on the &lt;a href=&#34;https://highlightjs.org/static/demo/&#34; target=&#34;_blank&#34;&gt;highlight.js demo page&lt;/a&gt; are available from the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs server&lt;/a&gt;. If you want to use a style that is not served by cdnjs, just leave &lt;code&gt;highlight_style&lt;/code&gt; unset, and place the corresponding css file in &lt;code&gt;/static/css/highlight.min.css&lt;/code&gt;.
Missing &lt;strong&gt;alert&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;If you don&amp;rsquo;t want to change the default style that ships with Academic but you do want the style file served from the &lt;a href=&#34;https://cdnjs.com/libraries/highlight.js/&#34; target=&#34;_blank&#34;&gt;cdnjs server&lt;/a&gt;, set &lt;code&gt;highlight_style = &amp;quot;github&amp;quot;&lt;/code&gt; in &lt;code&gt;config.toml&lt;/code&gt;.
Missing &lt;strong&gt;alertend&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;The &lt;code&gt;highlight_style&lt;/code&gt; option is only recognized when set in &lt;code&gt;config.toml&lt;/code&gt;. Setting &lt;code&gt;highlight_style&lt;/code&gt; in your page&amp;rsquo;s preamble has no effect.&lt;/p&gt;

&lt;h4 id=&#34;option-highlight-version&#34;&gt;Option &lt;code&gt;highlight_version&lt;/code&gt;&lt;/h4&gt;

&lt;p&gt;The &lt;code&gt;highlight_version&lt;/code&gt; option, as the name implies, allows you to select the version of highlight.js you want to use. The default value is &amp;ldquo;9.9.0&amp;rdquo;. The &lt;code&gt;highlight_version&lt;/code&gt; option is only recognized when set in &lt;code&gt;config.toml&lt;/code&gt;. Setting &lt;code&gt;highlight_version&lt;/code&gt; in your page&amp;rsquo;s preamble has no effect.&lt;/p&gt;

&lt;h2 id=&#34;twitter-tweet&#34;&gt;Twitter tweet&lt;/h2&gt;

&lt;p&gt;To include a single tweet, pass the tweet’s ID from the tweet&amp;rsquo;s URL as parameter to the shortcode:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; tweet 666616452582129664 &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;youtube&#34;&gt;Youtube&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; youtube w7Ft2ymGmfc &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;vimeo&#34;&gt;Vimeo&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; vimeo 146022717 &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;github-gist&#34;&gt;GitHub gist&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; gist USERNAME GIST-ID  &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;speaker-deck&#34;&gt;Speaker Deck&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;{{&amp;lt; speakerdeck 4e8126e72d853c0060001f97 &amp;gt;}}
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;rm-latex-math&#34;&gt;$\rm \LaTeX$ math&lt;/h2&gt;

&lt;pre&gt;&lt;code class=&#34;language-TeX&#34;&gt;$$\left [ – \frac{\hbar^2}{2 m} \frac{\partial^2}{\partial x^2} + V \right ] \Psi = i \hbar \frac{\partial}{\partial t} \Psi$$
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;$$\left [ – \frac{\hbar^2}{2 m} \frac{\partial^2}{\partial x^2} + V \right ] \Psi = i \hbar \frac{\partial}{\partial t} \Psi$$&lt;/p&gt;

&lt;p&gt;Alternatively, inline math can be written by wrapping the formula with only a single &lt;code&gt;$&lt;/code&gt;:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;This is inline: $\mathbf{y} = \mathbf{X}\boldsymbol\beta + \boldsymbol\varepsilon$
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;This is inline: $\mathbf{y} = \mathbf{X}\boldsymbol\beta + \boldsymbol\varepsilon$&lt;/p&gt;

&lt;h2 id=&#34;table&#34;&gt;Table&lt;/h2&gt;

&lt;p&gt;Code:&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-Markdown&#34;&gt;| Command           | Description                    |
| ------------------| ------------------------------ |
| `hugo`            | Build your website.            |
| `hugo serve -w`   | View your website.             |
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;Result:&lt;/p&gt;

&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Command&lt;/th&gt;
&lt;th&gt;Description&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;

&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;hugo&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Build your website.&lt;/td&gt;
&lt;/tr&gt;

&lt;tr&gt;
&lt;td&gt;&lt;code&gt;hugo serve -w&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;View your website.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;

&lt;h2 id=&#34;alerts&#34;&gt;Alerts&lt;/h2&gt;

&lt;p&gt;Alerts are a useful feature that add side content such as tips, notes, or warnings to your articles. They are especially handy when writing educational tutorial-style articles. Use the corresponding shortcodes to enable alerts inside your content.
Missing &lt;em&gt;alert&lt;/em&gt; templates:&lt;/p&gt;

&lt;p&gt;This will display the following &lt;em&gt;note&lt;/em&gt; block:&lt;/p&gt;

&lt;p&gt;This will display the following &lt;em&gt;warning&lt;/em&gt; block:&lt;/p&gt;

&lt;p&gt;Here&amp;rsquo;s some important information&amp;hellip;&lt;/p&gt;
&lt;div class=&#34;footnotes&#34;&gt;

&lt;hr /&gt;

&lt;ol&gt;
&lt;li id=&#34;fn:1&#34;&gt;Footnote example.
 &lt;a class=&#34;footnote-return&#34; href=&#34;#fnref:1&#34;&gt;&lt;sup&gt;^&lt;/sup&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ol&gt;
&lt;/div&gt;</description>
    </item>

    <item>
      <title>Manifold matching: Joint optimization of fidelity and commensurability</title>
      <link>https://adalisan.github.io/publication/manmatch-jofc/</link>
      <pubDate>Tue, 01 Sep 2015 00:00:00 +0000</pubDate>

      <guid>https://adalisan.github.io/publication/manmatch-jofc/</guid>
      <description></description>
    </item>

  </channel>
</rss>