User Tools

Site Tools


projects:hope:start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
projects:hope:start [2015/02/12 02:51] hjprojects:hope:start [2016/04/25 14:55] (current) hj
Line 7: Line 7:
 Each hidden layer in neural networks can be formulated as one HOPE model [1] Each hidden layer in neural networks can be formulated as one HOPE model [1]
 \\ \\
-  * The HOPE model combines a linear orthogonal projection and a mixture model under a uni ed generative modelling framework; +  * The HOPE model combines a linear orthogonal projection and a mixture model under a uni fied generative modelling framework; 
   * The HOPE model can be used as a novel tool to probe why and how NNs work;    * The HOPE model can be used as a novel tool to probe why and how NNs work; 
-  * The HOPE model provides several new learning algorithms to learn NNs either supervisedly or unsupervisedly+  * The HOPE model provides several new learning algorithms to learn NNs in either supervised or unsupervised ways
  
 **Reference:**  **Reference:** 
 \\ \\
 [1] //Shiliang Zhang and Hui Jiang//, "Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks," [[http://arxiv.org/abs/1502.00702|arXiv:1502.00702]].  [1] //Shiliang Zhang and Hui Jiang//, "Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks," [[http://arxiv.org/abs/1502.00702|arXiv:1502.00702]]. 
 +
 +[2] //Shiliang Zhang, Hui Jiang, Lirong Dai//, "The New HOPE Way to Learn Neural Networks", Proc. of Deep Learning Workshop at ICML 2015, July 2015. ([[https://8109f4a4-a-62cb3a1a-s-sites.googlegroups.com/site/deeplearning2015/9.pdf?attachauth=ANoY7coK1o0yBn1dsWEI0HgzM8TRfaBZAVw5BlivHyIiOJt3Hz1Q-fSv3qKwTo4eeS2bdcIrseorD-GCFaggMP9n6EKExhoYnm8smj5LogA6kmlM5SPmMWJxf3uhMRKk5T0M9_cEdtCVwCBSQh4Ya3PYW5OogwFzibs-vpFIZfN6OQG50CmwNBK4x3-tyi0nrihKfNFeMMbUyMWZeeSj9ToElpkc2qB17A%3D%3D&attredirects=0|paper]])
 +
 +[3] //S. Zhang, H. Jiang, L. Dai//, “[[http://jmlr.org/papers/v17/15-335.html| Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Learn Neural Networks]],” //Journal of Machine Learning Research (JMLR)//, 17(37):1−33, 2016.
  
 **Software:** **Software:**
 \\ \\
-The matlab codes to reproduce the MNIST results in [1] can be downloaded here.+The matlab codes to reproduce the MNIST results in [1,2] can be downloaded {{:projects:hope:hope4mnist_matlab.rar|here}}. 
 + 
 +<HTML> 
 +<script> 
 +  (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ 
 +  (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), 
 +  m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) 
 +  })(window,document,'script','//www.google-analytics.com/analytics.js','ga'); 
 + 
 +  ga('create', 'UA-59662420-1', 'auto'); 
 +  ga('send', 'pageview');
  
 +</script>
 +</HTML>
projects/hope/start.1423709476.txt.gz · Last modified: 2015/02/12 02:51 by hj

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki