Difference between revisions of "Active Learning"
From WikiBiron
Line 3:  Line 3:  
<ul>  <ul>  
−  <  +  <li>A. Gadde, A. Anis and A. Ortega "[http://sipi.usc.edu/~ortega/Papers/Gadde_KDD_14.pdf Active SemiSupervised Learning Using Sampling Theory for Graph Signals]", <i>KDD 2014</i>, New York, USA, 2014. 
<LI>A. Anis, A. Gadde and A. Ortega "[http://sipi.usc.edu/~ortega/Papers/Anis_ICASSP_14.pdf Towards a Sampling Theorem for Signals on Arbitrary Graphs]", <i>ICASSP 2014</i>, Florence, Italy, 2014 (Best student paper award).  <LI>A. Anis, A. Gadde and A. Ortega "[http://sipi.usc.edu/~ortega/Papers/Anis_ICASSP_14.pdf Towards a Sampling Theorem for Signals on Arbitrary Graphs]", <i>ICASSP 2014</i>, Florence, Italy, 2014 (Best student paper award).  
Line 19:  Line 19:  
<ul>  <ul>  
−  The code  +  The code below applies the proposed active learning method to the USPS Handwritten Digit Recognition dataset. It is written in MATLAB R2013a. It uses the SGWT toolbox which is also included. To run the code, simply unpack the directory and run main_usps.m. If you have any questions, please email agadde at usc dot edu. 
+  
+  <\ul>  
[http://sipi.usc.edu/~ortega/Papers/activesslviasamplingtheorycode.zip Matlab Code]  [http://sipi.usc.edu/~ortega/Papers/activesslviasamplingtheorycode.zip Matlab Code] 
Revision as of 08:27, 21 January 2015
 A. Gadde, A. Anis and A. Ortega "Active SemiSupervised Learning Using Sampling Theory for Graph Signals", KDD 2014, New York, USA, 2014.
 A. Anis, A. Gadde and A. Ortega "Towards a Sampling Theorem for Signals on Arbitrary Graphs", ICASSP 2014, Florence, Italy, 2014 (Best student paper award).
 S.K. Narang, A. Gadde and A. Ortega "Signal Processing Techniques for Interpolation of Graph Structured Data", ICASSP 2013, Vancouver, Canada, 2013.
 S.K. Narang, A. Gadde, E. Sanou and A. Ortega "Localized Iterative Methods for Interpolation of Graph Structured Data", GlobalSIP 2013, Austin, USA, 2013.

The code below applies the proposed active learning method to the USPS Handwritten Digit Recognition dataset. It is written in MATLAB R2013a. It uses the SGWT toolbox which is also included. To run the code, simply unpack the directory and run main_usps.m. If you have any questions, please email agadde at usc dot edu.
<\ul>
Matlab Code