<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:CMR12;
panose-1:0 0 0 0 0 0 0 0 0 0;}
@font-face
{font-family:SFRM1200;
panose-1:0 0 0 0 0 0 0 0 0 0;}
@font-face
{font-family:"\@CMR12";
panose-1:0 0 0 0 0 0 0 0 0 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="#0563C1" vlink="#954F72" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">Instructors: Kyle Shiflett and Avinash Karanth<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">Credit Hours: 3<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">EE 4900/5900 is intended to introduce students to basic deep neural networks, and provide an indepth<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">study of computer architecture methods for efficient training and inference of deep neural<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">networks. The recent proliferation of artificial intelligence, in particular deep neural networks<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">(DNNs), has led to an increasing pressure on hardware systems that run these models. DNNs<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">have established some of the leading state-of-the-art models for tasks such as image classification<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">and speech recognition, some even achieving super-human accuracy. As the size and complexity<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">of DNN models continue to grow, as does the need for energy-efficient and fast execution of these<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">workloads. This course focuses on recent computer architecture trends and hardware-software<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">co-design techniques that facilitate efficient execution of DNNs.<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:CMR12">Topics covered in this course include:<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Multilayer perceptrons<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Convolutional neural networks<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Affine and nonlinear integer quantization<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Operand dataflow and stationarity<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Hardware accelerators<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Compression with sparsity and pruning<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Memory organization<o:p></o:p></span></p>
<p class="MsoNormal" style="text-autospace:none"><span style="font-size:12.0pt;font-family:SFRM1200">•
</span><span style="font-size:12.0pt;font-family:CMR12">Interconnects<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:12.0pt;font-family:SFRM1200">• </span>
<span style="font-size:12.0pt;font-family:CMR12">Training</span><o:p></o:p></p>
</div>
</body>
</html>