-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy paththis-week-in-artificial-intelligence-security-34-of-2016.html
192 lines (169 loc) · 9.79 KB
/
this-week-in-artificial-intelligence-security-34-of-2016.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>This week in Artificial Intelligence & Security - #34 of 2016</title>
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta name="description" content="">
<meta name="author" content="Marina von Steinkirch">
<!-- Le styles -->
<link rel="stylesheet" href="./theme/css/bootstrap.dark.css" type="text/css" />
<style type="text/css">
body {
padding-top: 60px;
padding-bottom: 40px;
}
.tag-1 {
font-size: 13pt;
}
.tag-2 {
font-size: 11pt;
}
.tag-2 {
font-size: 10pt;
}
.tag-4 {
font-size: 8pt;
}
</style>
<link href="./theme/css/bootstrap-responsive.dark.css" rel="stylesheet">
<link href="./theme/css/font-awesome.css" rel="stylesheet">
<link href="./theme/css/pygments.css" rel="stylesheet">
<!-- Le fav and touch icons -->
<link rel="shortcut icon" href="./theme/images/favicon.ico">
<link rel="apple-touch-icon" href="./theme/images/apple-touch-icon.png">
<link rel="apple-touch-icon" sizes="72x72" href="./theme/images/apple-touch-icon-72x72.png">
<link rel="apple-touch-icon" sizes="114x114" href="./theme/images/apple-touch-icon-114x114.png">
<link href="./feeds/all.atom.xml" type="application/atom+xml" rel="alternate" title="chmod +x singularity.sh ATOM Feed" />
</head>
<body>
<div class="navbar navbar-fixed-top">
<div class="navbar-inner">
<div class="container-fluid">
<a class="btn btn-navbar" data-toggle="collapse" data-target=".nav-collapse">
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</a>
<a class="brand" href="./index.html">chmod +x singularity.sh </a>
<div class="nav-collapse">
<ul class="nav">
<li class="divider-vertical"></li>
<ul class="nav pull-right">
<li><a href="./authors.html">About</a></li>
<li><a href="./archives.html"><b>Archives</b></a></li>
<li>
<a href="https://github.com/bt3gl">github
<!--<i class="icon-github-sign icon-large" ></i>-->
</a></li>
<li>
<a href="https://twitter.com/1bt337">
<!--<i class="icon-twitter-sign icon-large"></i> -->
twitter
</a></li>
<li><a href="http://bt3gl.github.io/projects_page/index.html">Bygone Playful Times
</a></li>
</ul>
</ul>
<!--<p class="navbar-text pull-right">Logged in as <a href="#">username</a></p>-->
</div><!--/.nav-collapse -->
</div>
</div>
</div>
<div class="container-fluid">
<div class="row">
<div class="span9" id="content">
<section id="content">
<article>
<header>
<h1>
<a href=""
rel="bookmark"
title="Permalink to This week in Artificial Intelligence & Security - #34 of 2016">
This week in Artificial Intelligence & Security - #34 of 2016
</a>
</h1>
</header>
<div class="entry-content">
<div class="well">
<footer class="post-info">
<abbr class="published" title="2016-08-27T00:00:00">
Sat 27 August 2016 </abbr>
<span class="label"> Category</span>
<a href="./category/ai-ml-security.html"><i class="icon-folder-open"></i>AI & ML, Security</a>
</footer><!-- /.post-info --> </div>
<p><img alt="cyber" height="300px" src="./cyberpunk/7.gif" width="400px" /></p>
<h2>Papers</h2>
<ul>
<li><a href="https://arxiv.org/abs/1608.05343">Decoupled Neural Interfaces using Synthetic Gradients</a>. New DeepMind's paper reviewing backpropagation, using a modeled synthetic gradient in place of true backpropagated error gradients. Backpropagation is a a bottleneck in DNN, so the idea is instead use an asynchronous estimator, that would be obtained by supervised training of another mini neural network (a fully-connected NN regressor with 0-3 layers). Some speedup was showed in DeepMind’s hardware, but still raising questions on whether this would add any improvement on low-end GPUs setup.</li>
<li><a href="http://arxiv.org/pdf/1608.05148v1.pdf">Full Resolution Image Compression with Recurrent
Neural Networks</a>.</li>
<li><a href="http://arxiv.org/pdf/1608.03983v2.pdf">SGDR: Stochastic Gradient Descent with Restarts</a>.</li>
<li><a href="http://arxiv.org/pdf/1608.04980v1.pdf">Mollifying Networks</a>.</li>
<li><a href="http://ai.stanford.edu/~nilsson/QAI/qai.pdf">The quest for Artificial Intelligence</a>.</li>
</ul>
<h4>Good'n'old</h4>
<ul>
<li><a href="http://arxiv.org/abs/1407.7906">How Auto-Encoders Could Provide Credit Assignment in Deep Networks via Target Propagation</a>.</li>
<li><a href="https://arxiv.org/pdf/1605.02026.pdf">Training Neural Networks Without Gradients:
A Scalable ADMM Approach</a>.</li>
<li><a href="http://arxiv.org/abs/1506.08473">Beating the Perils of Non-Convexity: Guaranteed Training of Neural Networks using Tensor Methods</a>.</li>
<li><a href="https://arxiv.org/pdf/1506.02025.pdf">Spatial Transformer Networks</a>.</li>
<li><a href="https://arxiv.org/pdf/1412.2306v2.pdf">Deep Visual-Semantic Alignments for Generating Image Descriptions</a>.</li>
<li><a href="https://arxiv.org/pdf/1406.2661v1.pdf">Generative Adversarial Nets</a>.</li>
</ul>
<h2>News</h2>
<ul>
<li><a href="http://www.sixthtone.com/news/weiqi-players-go-flow">‘Weiqi’ Players Go With the Flow</a>.</li>
<li><a href="https://medium.com/@philjama/how-tensors-advance-human-technology-3831bff0906#.mp19g9px5">How Tensors Advance Human Technology</a>.</li>
<li><a href="https://pemami4911.github.io/blog_posts/2016/08/21/ddpg-rl.html">Deep Deterministic Policy Gradients in TensorFlow</a>.</li>
<li><a href="http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/">RNNs in Tensorflow, a Practical Guide</a>.</li>
<li><a href="https://medium.com/@ivydatascience/landscape-of-deep-learning-frameworks-aae34564cab#.m9r6p2uvs">Landscape of Deep Learning Frameworks</a>.</li>
<li><a href="https://devblogs.nvidia.com/parallelforall/deep-learning-self-driving-cars/?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=revue#.V7hqFpKnLPo.google_plusone_share">End-to-End Deep Learning for Self-Driving Cars</a>.</li>
<li><a href="http://nlpers.blogspot.com/2016/08/debugging-machine-learning.html">Debugging machine learning</a>.</li>
<li><a href="https://research.googleblog.com/2016/08/text-summarization-with-tensorflow.html">Text summarization with TensorFlow</a>.</li>
<li><a href="http://blog.aylien.com/introduction-generative-adversarial-networks-code-tensorflow/">An introduction to Generative Adversarial Networks (with code in TensorFlow)</a>.</li>
<li><a href="http://deliprao.com/archives/191">Is BackPropagation Necessary?</a>.</li>
<li><a href="http://blog.fastforwardlabs.com/post/149329060653/under-the-hood-of-the-variational-autoencoder-in">Under the Hood of the Variational Autoencoder (in Prose and Code)</a>.</li>
<li><a href="https://rubenfiszel.github.io/posts/rl4j/2016-08-24-Reinforcement-Learning-and-DQN.html">Reinforcement Learning and DQN, learning to play from pixels</a>.</li>
</ul>
<h2>Tools</h2>
<ul>
<li><a href="https://github.com/facebookresearch/fastText">fastText, Library for fast text representation and classification.</a>.</li>
<li><a href="https://github.com/bamos/dcgan-completion.tensorflow?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=revue">Image Completion with Deep Learning in TensorFlow</a>.</li>
</ul>
<h2>Videos</h2>
<ul>
<li><a href="https://www.youtube.com/watch?v=GccsFBQm-d4">TensorFlow: Demystifying Deep Learning with Visualizations</a>.</li>
<li><a href="https://www.youtube.com/watch?v=ohmajJTcpNk">Face2Face: Real-time Face Capture and Reenactment of RGB Videos (CVPR 2016 Oral)</a>.</li>
<li><a href="https://www.youtube.com/watch?v=2FmcHiLCwTU">TensorFlow in 5 Minutes</a>.</li>
</ul>
<p><img alt="cyber" height="500px" src="./draws/a1.png" width="600px" /></p>
</div><!-- /.entry-content -->
<div class="comments">
<h2>Comments !</h2>
<div id="disqus_thread"></div>
<script type="text/javascript">
var disqus_identifier = "this-week-in-artificial-intelligence-security-34-of-2016.html";
(function() {
var dsq = document.createElement('script');
dsq.type = 'text/javascript'; dsq.async = true;
dsq.src = 'http://bt3gl.disqus.com/embed.js';
(document.getElementsByTagName('head')[0] ||
document.getElementsByTagName('body')[0]).appendChild(dsq);
})();
</script>
</div>
</article>
</section>
</div><!--/span-->
</div><!--/row-->
<footer>
<address id="about">
</address><!-- /#about -->
</footer>
</div><!--/.fluid-container-->
<script src="./theme/js/jquery-1.7.2.min.js"></script>
<script src="./theme/js/bootstrap.min.js"></script>
</body>
</html>