update links to Github HW

_AIbiomedPage/AIbiomed.md

@@ -62,7 +62,7 @@ the field of deep learning has lead to groundbreaking performance in many applic
 
 
 ## Contacts:
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage/).
 
 Thanks for reading!
 

_AIbiomedPage/fastsk-archive-grid.md

@@ -51,6 +51,6 @@ Existing gkm-SVM algorithms
 
 
 ## Contact
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage/).
 
 Thanks for reading!

_AIfastConnectomePage/AIfastConnectome.md

@@ -85,7 +85,7 @@ We have designed a suite of novel and fast machine-learning algorithms to identi
 
 
 ## Contact
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage//).
 
 Thanks for reading!
 

_AIselfPage/AIself-archive-grid.md

@@ -50,7 +50,7 @@ Developing such techniques are an active research area. We focus on investigatin
 
 
 ## Contacts:
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage/).
 
 Thanks for reading!
 

_AItrustPage/AItrust-archive-grid.md

@@ -51,7 +51,7 @@ At the junction between NLP, deep learning and computer security, we build toolb
 
 
 ## Contact
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage/).
 
 Thanks for reading!
 

_AqgroupPage/gmember.htm

@@ -17,8 +17,8 @@
 
 
     <ul>
-    	<li>Mentor: <a href="http://www.cs.virginia.edu/yanjun/">Yanjun Qi</a> (Faculty at
-         <a href="http://www.cs.virginia.edu/people/faculty/qi.html">
+    	<li>Mentor: <a href="https://qiyanjun.github.io/Homepage/">Yanjun Qi</a> (Faculty at
+         <a href="https://qiyanjun.github.io/Homepage/">
          Department of Computer Science</a>, in
                         <a href="http://www.cs.virginia.edu/">University
          of Virginia</a>)

_AqgroupPage/gnews.htm

@@ -252,7 +252,7 @@ <h2>What's New<br></h2>
 
 <li><font face="Arial,Helvetica">[2016/09/16]:  <font face="Arial,Helvetica">
   Dr. Qi, gave an invited talk @ UVA DSI
-  </font>  (<a href="https://dsi.virginia.edu/datapalooza/panelists">Datapalooza Event</a>). Talk Title: "Machine Learning for Big Data in Biomedicine".  (<a href="http://www.cs.virginia.edu/yanjun/paperA14/2016-QI-DatapaloozaOnline.pdf">Slide</a>)
+  </font>  (<a href="https://dsi.virginia.edu/datapalooza/panelists">Datapalooza Event</a>). Talk Title: "Machine Learning for Big Data in Biomedicine".  (<a href="https://qiyanjun.github.io/Homepage//paperA14/2016-QI-DatapaloozaOnline.pdf">Slide</a>)
 </li>
 
 
@@ -329,7 +329,7 @@ <h2>What's New<br></h2>
 
 
 <li>[2015/06/23]: Dr. Qi gave an invited talk about <a
-  href="http://www.cs.virginia.edu/yanjun/paperA14/201506-careerTalk-ITEHighSchool-online.pdf"> "Machine Learning
+  href="https://qiyanjun.github.io/Homepage//paperA14/201506-careerTalk-ITEHighSchool-online.pdf"> "Machine Learning
   and A Personal Journey for Engineering" </a> at 2015 <a
   href="http://www.seas.virginia.edu/admin/diversity/pre_college/ite.php">"UVA Introduction
   to Engineering (ITE) Program"</a> for pre-college high school students.
@@ -388,13 +388,13 @@ <h2>What's New<br></h2>
 
 <ul><div id="tagPI_tagServices_tagAward_14">
 <li>[2014/06/25]: Dr. Qi gave an invited talk about <a
-  href="http://www.cs.virginia.edu/yanjun/paperA14/201406-careerTalk-ITEHighSchool-online.pdf"> "Machine Learning
+  href="https://qiyanjun.github.io/Homepage//paperA14/201406-careerTalk-ITEHighSchool-online.pdf"> "Machine Learning
   and A Personal Journey for Engineering" </a> at 2014 <a
   href="http://www.seas.virginia.edu/admin/diversity/pre_college/ite.php">"UVA Introduction
   to Engineering (ITE) Program"</a> for pre-college high school students.
 </li>
 <li>[2014/03/05]: Dr. Qi gave an invited talk about <a
-  href="http://www.cs.virginia.edu/yanjun/paperA14/20140305-CHPG-Talk-online.pdf">
+  href="https://qiyanjun.github.io/Homepage//paperA14/20140305-CHPG-Talk-online.pdf">
   "Machine Learning for Big Data Complexity in Biomedicine" </a> at
   2014 <a
   href="http://www.healthsystem.virginia.edu/events/index.cfm?viewtype=som&viewformat=event&eventinstanceid=31534"> "Genome Sciences Seminar Series"

_config.yml

@@ -96,7 +96,7 @@ author:
   links:
     - label: "HomePage"
       icon: "fas fa-fw fa-link"
-      url: "https://www.cs.virginia.edu/yanjun/"
+      url: "https://qiyanjun.github.io/Homepage/"
     - label: "Twitter"
       icon: "fab fa-fw fa-twitter-square"
       url: "https://twitter.com/Qdatalab"

_posts/2011-05-01-deepQA.md

@@ -31,7 +31,7 @@ In this notebook, we describe the automatic retrieval runs from NEC Laboratories
 
 
 ### Paper3: Kernelized information-theoretic metric learning for cancer diagnosis using high-dimensional molecular profiling data
-+ [PDF](https://www.cs.virginia.edu/yanjun/paperA14/2015-TKDD.pdf)
++ [PDF](https://qiyanjun.github.io/Homepage//paperA14/2015-TKDD.pdf)
 
 + Abstract
 With the advancement of genome-wide monitoring technologies, molecular expression data have become widely used for diagnosing cancer through tumor or blood samples. When mining molecular signature data, the process of comparing samples through an adaptive distance function is fundamental but difficult, as such datasets are normally heterogeneous and high dimensional. In this article, we present kernelized information-theoretic metric learning (KITML) algorithms that optimize a distance function to tackle the cancer diagnosis problem and scale to high dimensionality. By learning a nonlinear transformation in the input space implicitly through kernelization, KITML permits efficient optimization, low storage, and improved learning of distance metric. We propose two novel applications of KITML for diagnosing cancer using high-dimensional molecular profiling data: (1) for sample-level cancer diagnosis, the learned metric is used to improve the performance of k-nearest neighbor classification; and (2) for estimating the severity level or stage of a group of samples, we propose a novel set-based ranking approach to extend KITML. For the sample-level cancer classification task, we have evaluated on 14 cancer gene microarray datasets and compared with eight other state-of-the-art approaches. The results show that our approach achieves the best overall performance for the task of molecular-expression-driven cancer sample diagnosis. For the group-level cancer stage estimation, we test the proposed set-KITML approach using three multi-stage cancer microarray datasets, and correctly estimated the stages of sample groups for all three studies.

_posts/2013-03-08-OldSummaryTalk.md

@@ -8,7 +8,7 @@ tag:
 Here are the slides of one lecture talk I gave at UVA [CPHG](https://med.virginia.edu/cphg/) Seminar Series in 2014 about our deep learning tools back then. 
 
 
-### Slides: [@URL](http://www.cs.virginia.edu/yanjun/paperA14/20140305-CHPG-Talk-online.pdf)
+### Slides: [@URL](https://qiyanjun.github.io/Homepage/paperA14/20140305-CHPG-Talk-online.pdf)
 
 
 

_posts/2014-06-20-DeepSparseCoding.md

@@ -14,7 +14,7 @@ tag:
 
 
 
-### [Talk PDF](http://www.cs.virginia.edu/yanjun/paperA14/2014-sdm-deepsc-talk.pdf)
+### [Talk PDF](https://qiyanjun.github.io/Homepage//paperA14/2014-sdm-deepsc-talk.pdf)
 
 
 ### Abstract:

_posts/2014-06-20-latentDependency.md

@@ -16,7 +16,7 @@ tag:
 
 ### [GitHub](https://github.com/DeepLearning4BioSeqText/Paper12-NIPS-SparseGGM4LatentFactors)
 
-### [Poster PDF](https://www.cs.virginia.edu/yanjun/paperA14/2012_SLFA_NIPS.pdf)
+### [Poster PDF](https://qiyanjun.github.io/Homepage//paperA14/2012_SLFA_NIPS.pdf)
 
 
 

_posts/2016-06-01-fasjem.md

@@ -25,7 +25,7 @@ demo(fasjem)
 
 ### Talk [URL](https://github.com/QData/FASJEM/blob/master/17-FASJEM-talk.pdf)
 
-### [Poster](http://www.cs.virginia.edu/yanjun/paperA14/2017-aistat-poster-simule.pdf)
+### [Poster](https://qiyanjun.github.io/Homepage//paperA14/2017-aistat-poster-simule.pdf)
 
 ### Abstract
 Estimating multiple sparse Gaussian Graphical

_posts/2017-05-11-Theory.md

@@ -8,7 +8,7 @@ tag:
 
 ### Paper [ICLR17 workshop](https://arxiv.org/abs/1612.00334)
 
-### [Poster](http://www.cs.virginia.edu/yanjun/paperA14/2017-ICLR-poster-unified.pdf)
+### [Poster](https://qiyanjun.github.io/Homepage//paperA14/2017-ICLR-poster-unified.pdf)
 
 ### Abstract
 Most machine learning classifiers, including deep neural networks, are vulnerable to adversarial examples. Such inputs are typically generated by adding small but purposeful modifications that lead to incorrect outputs while imperceptible to human eyes. The goal of this paper is not to introduce a single method, but to make theoretical steps towards fully understanding adversarial examples. By using concepts from topology, our theoretical analysis brings forth the key reasons why an adversarial example can fool a classifier (f1) and adds its oracle (f2, like human eyes) in such analysis. By investigating the topological relationship between two (pseudo)metric spaces corresponding to predictor f1 and oracle f2, we develop necessary and sufficient conditions that can determine if f1 is always robust (strong-robust) against adversarial examples according to f2. Interestingly our theorems indicate that just one unnecessary feature can make f1 not strong-robust, and the right feature representation learning is the key to getting a classifier that is both accurate and strong-robust.

_posts/2017-06-01-Evade.md

@@ -8,15 +8,15 @@ tag:
 <a name="genetic"></a>
 
 ### Paper: Automatically Evading Classifiers, 
-A Case Study on PDF Malware Classifiers [NDSS16](http://www.cs.virginia.edu/yanjun/paperA14/2016-evade_classifier.pdf)
+A Case Study on PDF Malware Classifiers [NDSS16](https://qiyanjun.github.io/Homepage//paperA14/2016-evade_classifier.pdf)
 
 [More information is provided by EvadeML.org](http://evademl.org/)
 
 By using evolutionary techniques to simulate an adversary's efforts to evade that classifier
 
 ### GitHub: [EvadePDFClassifiers](https://github.com/uvasrg/EvadeML)
 
-### [Presentation](http://www.cs.virginia.edu/yanjun/paperA14/2016-evade-ndsst.pdf)
+### [Presentation](https://qiyanjun.github.io/Homepage//paperA14/2016-evade-ndsst.pdf)
 
 
 

_posts/2017-06-03-Defend.md

@@ -11,7 +11,7 @@ tag:
 
 ### GitHub: [DeepCloak](https://github.com/qdata/deepcloak)
 
-### [Poster](http://www.cs.virginia.edu/yanjun/paperA14/2017-ICLRposter_deepCloak.pdf)
+### [Poster](https://qiyanjun.github.io/Homepage//paperA14/2017-ICLRposter_deepCloak.pdf)
 
 ### Abstract
 Recent studies have shown that deep neural networks (DNN) are vulnerable to adversarial samples: maliciously-perturbed samples crafted to yield incorrect model outputs. Such attacks can severely undermine DNN systems, particularly in security-sensitive settings. It was observed that an adversary could easily generate adversarial samples by making a small perturbation on irrelevant feature dimensions that are unnecessary for the current classification task. To overcome this problem, we introduce a defensive mechanism called DeepCloak. By identifying and removing unnecessary features in a DNN model, DeepCloak limits the capacity an attacker can use generating adversarial samples and therefore increase the robustness against such inputs. Comparing with other defensive approaches, DeepCloak is easy to implement and computationally efficient. Experimental results show that DeepCloak can increase the performance of state-of-the-art DNN models against adversarial samples.

_posts/2017-06-12-Genome-MeMo.md

@@ -19,7 +19,7 @@ tag:
 
 ### [GitHub](https://github.com/QData/DeepMotif)
 
-### [Poster](http://www.cs.virginia.edu/yanjun/paperA14/2017-ICLR-Poster-MeMo.pdf)
+### [Poster](https://qiyanjun.github.io/Homepage//paperA14/2017-ICLR-Poster-MeMo.pdf)
 
 ### Abstract
 When analyzing the genome, researchers have discovered that proteins bind to DNA based on certain patterns of the DNA sequence known as "motifs". However, it is difficult to manually construct motifs due to their complexity. Recently, externally learned memory models have proven to be effective methods for reasoning over inputs and supporting sets. In this work, we present memory matching networks (MMN) for classifying DNA sequences as protein binding sites. Our model learns a memory bank of encoded motifs, which are dynamic memory modules, and then matches a new test sequence to each of the motifs to classify the sequence as a binding or nonbinding site.

_posts/2017-11-08-diffee.md

@@ -14,7 +14,7 @@ tag:
 
 ### Presentation: [Slides](https://github.com/QData/DIFFEE/blob/master/2018-DIFFEE-talk.pdf) @ AISTAT18
 
-### [Poster](http://www.cs.virginia.edu/yanjun/paperA14/2017-diffeenips17workshop.pdf) @  NIPS 2017 workshop for Advances in Modeling and Learning Interactions from Complex Data.
+### [Poster](https://qiyanjun.github.io/Homepage//paperA14/2017-diffeenips17workshop.pdf) @  NIPS 2017 workshop for Advances in Modeling and Learning Interactions from Complex Data.
 
 ### R package: [GitHub](https://github.com/QData/DIFFEE)
 

_posts/2018-08-19-TutorialsSample2Net.md

@@ -31,6 +31,6 @@ tag:
 
 
 ## Contact
-Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](http://www.cs.virginia.edu/yanjun/).
+Have questions or suggestions? Feel free to [ask me on Twitter](https://twitter.com/Qdatalab) or [email me](https://qiyanjun.github.io/Homepage//).
 
 Thanks for reading!

_posts/2019-06-18-kDIFFNet.md

@@ -11,7 +11,7 @@ tag:
 
 ## Tool kDIFFNet: Adding Extra Knowledge in Scalable Learning of Sparse Differential Gaussian Graphical Models
 
-### Paper: [BioArxiv](http://biorxiv.org/cgi/content/short/716852v1) & [PDF](http://www.cs.virginia.edu/yanjun/paperA14/2019-kDiffNet.pdf)
+### Paper: [BioArxiv](http://biorxiv.org/cgi/content/short/716852v1) & [PDF](https://qiyanjun.github.io/Homepage//paperA14/2019-kDiffNet.pdf)
 
 ### Abstract
 We focus on integrating different types of extra knowledge (other than the observed samples) for estimating the sparse structure change between two p-dimensional Gaussian Graphical Models (i.e. differential GGMs). Previous differential GGM estimators either fail to include additional knowledge or cannot scale up to a high-dimensional (large p) situation. This paper proposes a novel method KDiffNet that incorporates Additional Knowledge in identifying Differential Networks via an Elementary Estimator. We design a novel hybrid norm as a superposition of two structured norms guided by the extra edge information and the additional node group knowledge. KDiffNet is solved through a fast parallel proximal algorithm, enabling it to work in large-scale settings. KDiffNet can incorporate various combinations of existing knowledge without re-designing the optimization. Through rigorous statistical analysis we show that, while considering more evidence, KDiffNet achieves the same convergence rate as the state-of-the-art. Empirically on multiple synthetic datasets and one real-world fMRI brain data, KDiffNet significantly outperforms the cutting edge baselines concerning the prediction performance, while achieving the same level of time cost or less.