<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.9.2">Jekyll</generator><link href="gwding.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="gwding.github.io/" rel="alternate" type="text/html" /><updated>2022-07-04T00:39:17+00:00</updated><id>gwding.github.io/feed.xml</id><title type="html">Gavin Weiguang Ding (丁伟光)</title><subtitle>ML/CV Researcher and Engineer</subtitle><entry><title type="html">AdverTorch: A toolbox for adversarial robustness research</title><link href="gwding.github.io/advertorch-a-toolbox-for-adversarial-robustness-research/" rel="alternate" type="text/html" title="AdverTorch: A toolbox for adversarial robustness research" /><published>2019-04-30T00:00:00+00:00</published><updated>2019-04-30T00:00:00+00:00</updated><id>gwding.github.io/advertorch-a-toolbox-for-adversarial-robustness-research</id><content type="html" xml:base="gwding.github.io/advertorch-a-toolbox-for-adversarial-robustness-research/">&lt;p&gt;This is a short summary of the AdverTorch project that I lead. It is a Python toolbox for adversarial robustness research. The primary functionalities are implemented in PyTorch. Specifically, AdverTorch contains modules for generating adversarial perturbations and defending against adversarial examples, also scripts for adversarial training.&lt;/p&gt;

&lt;p&gt;The github repo is &lt;a href=&quot;https://github.com/BorealisAI/advertorch/&quot;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;The tech report is &lt;a href=&quot;https://arxiv.org/abs/1902.07623&quot;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;We also have a blog post &lt;a href=&quot;https://www.borealisai.com/en/blog/advertorch-adversarial-training-tool-implement-attack-and-defence-strategies/&quot;&gt;here&lt;/a&gt; on BorealisAI website.&lt;/p&gt;</content><author><name></name></author><summary type="html">This is a short summary of the AdverTorch project that I lead. It is a Python toolbox for adversarial robustness research. The primary functionalities are implemented in PyTorch. Specifically, AdverTorch contains modules for generating adversarial perturbations and defending against adversarial examples, also scripts for adversarial training.</summary></entry><entry><title type="html">Hello World</title><link href="gwding.github.io/Hello-World/" rel="alternate" type="text/html" title="Hello World" /><published>2015-12-27T00:00:00+00:00</published><updated>2015-12-27T00:00:00+00:00</updated><id>gwding.github.io/Hello-World</id><content type="html" xml:base="gwding.github.io/Hello-World/">&lt;p&gt;Welcome to my new website! It is still under construction.&lt;/p&gt;

&lt;p&gt;Thanks to github.io, &lt;a href=&quot;https://github.com/barryclark/jekyll-now&quot;&gt;Jekyll Now repository&lt;/a&gt; and disqus, I’m able to build this in an afternoon!&lt;/p&gt;</content><author><name></name></author><summary type="html">Welcome to my new website! It is still under construction.</summary></entry></feed>