<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Kind reminder of our meeting in 25'.
Please connect to<br>
<br>
<a class="moz-txt-link-freetext" href="https://bluejeans.com/788204027">https://bluejeans.com/788204027</a></div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">for your weekly dose of pyTorching.<br>
</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">On 2021-06-15 8:03 a.m., Eric Drechsler
wrote:<br>
</div>
<blockquote type="cite" cite="mid:590e644e-4e6c-be0f-9c68-8e14f38d0cc3@triumf.ca">
<p>Time to get schwifty!<br>
<br>
In our next meeting this <b>Wednesday, June 16th, 11am</b> we
will dissect the neural networks core material - no scalpel
needed. The focus will be on modules 2 & 3 -<b> Shallow and
Deep Neural Networks</b>, which include the topics:<br>
<br>
- one- & multidimensional NN, multiclass NN<br>
- backpropagation, activation functions<br>
- initialisation, momentum, dropout, batch normalisation<br>
<br>
All of these topics are rather important in practice, so please
make sure to bring any question which might arise during the
course review.<br>
<br>
We are looking for a <b>volunteer to guide the discussion</b>;
in [1] you can find the slides we used as guideline in our first
meeting. Our upcoming schedule is summarised in [2]. <br>
<br>
All hail Skynet,<br>
Your DS group<br>
<br>
[1]
<a class="moz-txt-link-freetext" href="https://docs.google.com/presentation/d/1x_s1-fsPsj9qMMqV1X-jscOU7i0cyL487hI_iSZx6rg/edit?usp=sharing" moz-do-not-send="true">https://docs.google.com/presentation/d/1x_s1-fsPsj9qMMqV1X-jscOU7i0cyL487hI_iSZx6rg/edit?usp=sharing</a><br>
[2]<br>
- June 16th, 11am: <br>
- Shallow Neural Networks<br>
- Deep Networks<br>
- June 23rd, 11am: <br>
- Convolutional NN<br>
- Autoencoders<br>
- June 30th: <br>
- Optional meeting<br>
- July 7th, 11am: <br>
- Peer Review <br>
- Final Exam<br>
</p>
</blockquote>
<p><br>
</p>
</body>
</html>