ODIn/Website
9
0
Fork 1
Code Pull requests Releases Wiki Activity

Turek talk

Browse source
This commit is contained in:
Oliver Kennedy 2018-12-06 22:45:45 -05:00
parent a4f533a270
commit d959ff9203
3 changed files with 7835 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

98
slides/talks/2018-6-Matt_Turek/index.html Normal file
View file

@ -0,0 +1,98 @@
<!doctype html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no">
<title>reveal.js</title>
<link rel="stylesheet" href="../../reveal.js-3.7.0/css/reveal.css">
<link rel="stylesheet" href="../../reveal.js-3.7.0/css/theme/white.css">
<!-- Theme used for syntax highlighting of code -->
<link rel="stylesheet" href="../../reveal.js-3.7.0/lib/css/zenburn.css">
<!-- Printing and PDF exports -->
<script>
var link = document.createElement( 'link' );
link.rel = 'stylesheet';
link.type = 'text/css';
link.href = window.location.search.match( /print-pdf/gi ) ? '../../reveal.js-3.7.0/css/print/pdf.css' : '../../reveal.js-3.7.0/css/print/paper.css';
document.getElementsByTagName( 'head' )[0].appendChild( link );
</script>
</head>
<body>
<div class="reveal"><div class="slides">
<section>
<h2>Quality-Aware Machine Learning</h2>
<p>Oliver Kennedy</p>
<p>Jaroslaw Zola</p>
<p>Matthew Knepley</p>
</section>
<section>
<svg data-src="ml-pipeline.svg" />
</section>
<section>
<section>
<h3>Fixing Data is Expensive</h3>
<p class="fragment">(or impossible)</p>
</section>
<section>
<h3>Re-using already fixed data is dangerous.</h3>
<p class="fragment">(the "right" fix depends on use case)</p>
</section>
</section>
<section>
<section>
<h3>Idea: Track Errors</h3>
<p>Incomplete Databases store <i>possibilities</i>, not just certainties.</p>
</section>
<section>
<svg data-src="tag-pipeline.svg" />
</section>
<section>
<h3>Goals</h3>
<ul>
<li class="fragment" style="margin-top: 20px;">Statistically rigorous techniques for training classifiers, neural networks on incomplete databases.</li>
<li class="fragment" style="margin-top: 20px;">Models incorporating incompleteness information.
<div style="font-size: 70%; color: grey; margin-left: 20px;">"I didn't have enough training data" should be an allowed prediction.</div>
</li>
<li class="fragment" style="margin-top: 20px;">Incompleteness as an assist for model debugging.
<div style="font-size: 70%; color: grey; margin-left: 20px;">Which errors have the biggest impact on a prediction? <br/>
Which errors best explain an incorrect prediction?</div>
</li>
</ul>
</section>
</div></div>
<script src="../../reveal.js-3.7.0/lib/js/head.min.js"></script>
<script src="../../reveal.js-3.7.0/js/reveal.js"></script>
<script>
// More info about config & dependencies:
// - https://github.com/hakimel/reveal.js#configuration
// - https://github.com/hakimel/reveal.js#dependencies
Reveal.initialize({
dependencies: [
{ src: '../../reveal.js-3.7.0/plugin/svginline/data-src-svg.js' },
{ src: '../../reveal.js-3.7.0/plugin/markdown/marked.js' },
{ src: '../../reveal.js-3.7.0/plugin/markdown/markdown.js' },
{ src: '../../reveal.js-3.7.0/plugin/notes/notes.js', async: true },
{ src: '../../reveal.js-3.7.0/plugin/highlight/highlight.js', async: true, callback: function() { hljs.initHighlightingOnLoad(); } }
]
});
</script>
</body>
</html>

3020
slides/talks/2018-6-Matt_Turek/ml-pipeline.svg Normal file
View file

File diff suppressed because it is too large Load diff
Side by side

After

Width:  |  Height:  |  Size: 138 KiB

4717
slides/talks/2018-6-Matt_Turek/tag-pipeline.svg Normal file
View file

File diff suppressed because it is too large Load diff
Side by side

After

Width:  |  Height:  |  Size: 216 KiB