“2020-07-13”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(7位用户的7个中间修订版本未显示)
第6行: 第6行:
 
|Dong Wang
 
|Dong Wang
 
||  
 
||  
*  
+
* deepnorm2 paper refinement.
 +
* More investigation on the optimum with NL, e.g., the global thresholding property
 +
* Formulation for AV denosing and phone-aware DNF/NDA
 +
* Continue the NB book.
 
||
 
||
*  
+
* Complete the formulation of the AV denoising approach.
 +
* Continue the NB book
 
||
 
||
 
*   
 
*   
第17行: 第21行:
 
|Yunqi Cai
 
|Yunqi Cai
 
||   
 
||   
*
+
*experiments on DNF2 and reconstructing paper structure
 
||
 
||
*  
+
* experiments on New DNF2 paper
 
||
 
||
 
*   
 
*   
第28行: 第32行:
 
|Zhiyuan Tang
 
|Zhiyuan Tang
 
||  
 
||  
*  
+
* DT + DNF Regularization.
 
||
 
||
*  
+
* Domian adaptation.
 
||
 
||
 
*   
 
*   
第39行: 第43行:
 
|Lantian Li
 
|Lantian Li
 
||  
 
||  
*  
+
* Enroll-test mismatch (test on large datasets)
 
||
 
||
*  
+
* Enroll-test mismatch.
||
+
+
 
|-
 
|-
  
第50行: 第52行:
 
|Ying Shi
 
|Ying Shi
 
||  
 
||  
* BP + length norm enhancement method
+
* Sub space flow plan A and plan B
* Sub-space flow
+
* visulization and analysis of Sub space r code
 
||
 
||
 
* Continue on Sub-space flow
 
* Continue on Sub-space flow
** Multi-SNR variance
 
 
** Cyclic loss
 
** Cyclic loss
* Sub-space plan B
+
* Re-train a DAE model with state of the art result
 
||
 
||
 
*   
 
*   
第65行: 第66行:
 
|Haoran Sun
 
|Haoran Sun
 
||  
 
||  
*  
+
* Analyzed latent space of VAE and AE on TIMIT
 
||
 
||
*  
+
* The same analysis on VAE with larger latent space dimension
 +
* The same analysis on Glow
 +
* Supervised approaches
 
||
 
||
 
*   
 
*   
第101行: 第104行:
 
|Ruiqi Liu
 
|Ruiqi Liu
 
||  
 
||  
*
+
*Complete the experiment and write the paper.
 
||
 
||
*
+
*Continue.
 
||
 
||
 
*   
 
*   
第111行: 第114行:
 
|Haolin Chen
 
|Haolin Chen
 
||  
 
||  
*  
+
* Subspace Glow test
 +
* Subspace Flow training and test
 
||
 
||
*  
+
* Subspace Flow planA, B, cycle loss
 
||
 
||
 
*   
 
*   

2020年7月19日 (日) 23:54的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • deepnorm2 paper refinement.
  • More investigation on the optimum with NL, e.g., the global thresholding property
  • Formulation for AV denosing and phone-aware DNF/NDA
  • Continue the NB book.
  • Complete the formulation of the AV denoising approach.
  • Continue the NB book
Yunqi Cai
  • experiments on DNF2 and reconstructing paper structure
  • experiments on New DNF2 paper
Zhiyuan Tang
  • DT + DNF Regularization.
  • Domian adaptation.
Lantian Li
  • Enroll-test mismatch (test on large datasets)
  • Enroll-test mismatch.
Ying Shi
  • Sub space flow plan A and plan B
  • visulization and analysis of Sub space r code
  • Continue on Sub-space flow
    • Cyclic loss
  • Re-train a DAE model with state of the art result
Haoran Sun
  • Analyzed latent space of VAE and AE on TIMIT
  • The same analysis on VAE with larger latent space dimension
  • The same analysis on Glow
  • Supervised approaches
Yue Fan
Jiawen Kang
  • Prepare new Near-far data
  • Collect experiments data
  • Write multi-conditon paper
  • Revise multi-conditon paper
  • MAML experiments
Ruiqi Liu
  • Complete the experiment and write the paper.
  • Continue.
Haolin Chen
  • Subspace Glow test
  • Subspace Flow training and test
  • Subspace Flow planA, B, cycle loss