fix hugo version

Author: vijaydwivedi75

404.html

@@ -640,7 +640,7 @@ <h2>Latest</h2>
 
 
   <p class="powered-by">
-    Xavier Bresson &copy; 2020 · Made with &hearts; by <a href="https://chaitjo.github.io/">Chaitanya Joshi</a> &middot; 
+    Xavier Bresson © 2020 · Made with ♥ by <a href="https://chaitjo.github.io/">Chaitanya Joshi</a> &middot; 
 
     Powered by the
     <a href="https://sourcethemes.com/academic/" target="_blank" rel="noopener">Academic theme</a> for

authors/chaitanya-joshi/index.xml

@@ -19,7 +19,6 @@
       <guid>https://graphdeeplearning.github.io/authors/vijay-dwivedi/</guid>
       <description>&lt;p&gt;Vijay Dwivedi is a first year PhD student in Machine Learning at NTU, Singapore supervised by Dr. Xavier Bresson.
 His primary interest is developing Deep Learning algorithms on graph-structured data and their applications to domains such as quantum chemistry, social networks, etc.&lt;/p&gt;
-
 &lt;p&gt;Before starting his PhD, Vijay worked with Dr. Bresson as a Research Assistant in the same lab. He has a background in Computer Science (BTech) from MNNIT Allahabad, where he explored the fields of Natural Language Processing and Multi-Modal Systems.&lt;/p&gt;
 </description>
     </item>
@@ -30,7 +29,6 @@ His primary interest is developing Deep Learning algorithms on graph-structured
       <pubDate>Fri, 12 Jun 2020 00:00:00 +0000</pubDate>
       <guid>https://graphdeeplearning.github.io/authors/chaitanya-joshi/</guid>
       <description>&lt;p&gt;Chaitanya Joshi is a Research Assistant under Dr. Xavier Bresson at NTU, Singapore. His current research focuses on the emerging field of Graph Deep Learning and its applications for Operations Research and Combinatorial Optimization.&lt;/p&gt;
-
 &lt;p&gt;He graduated from NTU in 2019 as the Valedictorian of his cohort with a BEng in Computer Science and a specialization in Artificial Intelligence. He is passionate about building data-driven solutions for real-world problems, and has 3+ years of experience doing the same at companies and research labs in Singapore and Switzerland. He has co-authored patent applications and research papers at top Machine Learning conferences such as NeurIPS and ICLR.&lt;/p&gt;
 </description>
     </item>
@@ -119,7 +117,6 @@ During his PhD, he was a visiting student in sketchX Lab, Queen Mary University
       <pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>
       <guid>https://graphdeeplearning.github.io/authors/axel-nilsson/</guid>
       <description>&lt;p&gt;Axel Nilsson is an exchange researcher student under Dr. Xavier Bresson at NTU, Singapore. His research project focuses on Spectral Graph Neural Networks and their transferability.&lt;/p&gt;
-
 &lt;p&gt;Axel is a Master’s student in at EPFL and obtained a BSc from the same school in 2017.&lt;/p&gt;
 </description>
     </item>
@@ -130,9 +127,7 @@ During his PhD, he was a visiting student in sketchX Lab, Queen Mary University
       <pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>
       <guid>https://graphdeeplearning.github.io/authors/david-low/</guid>
       <description>&lt;p&gt;David Low is currently a PhD student under School of Computer Science &amp;amp; Engineering, NTU supervised by Associate Professor Xavier Bresson. His current research focuses on Deep Learning and its applications for Natural Language Processing.&lt;/p&gt;
-
 &lt;p&gt;Before starting his PhD, he cofounded two startups and worked as a data scientist at Infocomm Development Authority, Singapore. In 2016, he represented Singapore and National University of Singapore (NUS) in Data Science Game at France and clinched top spot among teams from Asia and America.&lt;/p&gt;
-
 &lt;p&gt;Throughout his career, David has engaged in data science projects ranging from banking, telco, e-commerce to insurance industry. Some of his works including sales forecast modeling, mineral deposits prediction and process optimization had won him awards in several machine learning competitions. Earlier in his career, David was involved in research collaborations with Carnegie Mellon University (CMU) and Massachusetts Institute of Technology (MIT) on separate projects funded by National Research Foundation and SMART.&lt;/p&gt;
 </description>
     </item>

authors/index.xml

@@ -17,22 +17,16 @@
       <link>https://graphdeeplearning.github.io/project/sketches/</link>
       <pubDate>Tue, 14 Jan 2020 16:19:27 +0800</pubDate>
       <guid>https://graphdeeplearning.github.io/project/sketches/</guid>
-      <description>
-
-&lt;h2 id=&#34;representation-learning-for-sketches&#34;&gt;Representation Learning for Sketches&lt;/h2&gt;
-
+      <description>&lt;h2 id=&#34;representation-learning-for-sketches&#34;&gt;Representation Learning for Sketches&lt;/h2&gt;
 &lt;p&gt;Human beings have been creating free-hand sketches, &lt;em&gt;i.e.&lt;/em&gt;, drawings without precise instruments, since &lt;a href=&#34;https://en.wikipedia.org/wiki/Cave_painting&#34;&gt;time immemorial&lt;/a&gt;.
 Due to the popularity of touchscreen interfaces, machine learning using sketches has emerged as an interesting problem with a myriad of applications:
 If we consider sketches as 2D images, we can throw them into off-the-shelf &lt;a href=&#34;https://arxiv.org/abs/1501.07873&#34;&gt;Convolutional Neural Networks (CNNs)&lt;/a&gt;.
 While CNNs are designed for &lt;em&gt;static&lt;/em&gt; collections of pixels with &lt;em&gt;dense&lt;/em&gt; colors and textures,
 sketches are usually an extremely &lt;em&gt;sparse&lt;/em&gt; sequences of strokes which capture high-level abstractions and ideas. &lt;a href=&#34;https://ai.googleblog.com/2017/04/teaching-machines-to-draw.html&#34;&gt;Recurrent Neural Networks (RNNs)&lt;/a&gt; stick out as a natural architecture for capturing this temporal nature of sketches.&lt;/p&gt;
-
 &lt;blockquote&gt;
 &lt;p&gt;&lt;em&gt;Structure vs. temporal order: can we have the best of both worlds?&lt;/em&gt;&lt;/p&gt;
 &lt;/blockquote&gt;
-
 &lt;h2 id=&#34;sketches-as-graphs&#34;&gt;Sketches as Graphs&lt;/h2&gt;
-
 &lt;p&gt;We are working on a novel representation of free-hand sketches as &lt;strong&gt;sparsely-connected graphs&lt;/strong&gt;.
 We assume that sketches are sets of curves and strokes, which are discretized by a set of points representing the graph nodes.
 Each node encodes spatial, temporal and semantic information.