“NLP Status Report 2016-09-12”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(3位用户的5个中间修订版本未显示)
第36行: 第36行:
 
*understand the code of rnn grammar.
 
*understand the code of rnn grammar.
 
|-
 
|-
|Jiyuan Zhang || ||
+
|Jiyuan Zhang ||
 +
*wrote a report on the music generation [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/b/b4/%E5%9F%BA%E4%BA%8ERNN-RBM%E7%9A%84%E9%9F%B3%E4%B9%90%E7%94%9F%E6%88%90%E6%8A%A5%E5%91%8A.pdf here]]
 +
*completed the text section of my part of deep learning chapter,yet not completed the illustration [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/f/f3/%E7%AC%AC%E5%9B%9B%E7%AB%A0-%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0_-%E5%BC%A0%E8%AE%B0%E8%A2%81.pdf here]]
 +
||
 +
*complete the illustration
 +
*find some paper about generation task
 
|-
 
|-
 
|Aodong Li || ||
 
|Aodong Li || ||
 
|-
 
|-
 
| Andi Zhang || ||
 
| Andi Zhang || ||
 +
*look for source code of Memnn
 +
*prepare for the sharing of a paper
 
|-
 
|-
 
| Shiyao Li || ||
 
| Shiyao Li || ||
第49行: 第56行:
 
||
 
||
 
* Go on with the paper reading
 
* Go on with the paper reading
 +
* prepare for the bi-week report
 
|}
 
|}

2016年9月12日 (一) 05:30的最后版本

Date People Last Week This Week
2016/09/05 Yang Feng
  • surveyed memory networks -- papers and codes
  • surveyed the progress of Turing machine in 2016, mainly focusing on dynamic addressing Turing machine and reinforcement learning Turing machine.
  • implement the neural grammar based on memory networks
Jiyuan Zhang
  • started to write my part of deep learning chapter
  • perfected poem’s code according to qx’code
  • complete initial version of my part of book
  • generate the same level poem compared to qx’s
Aodong Li
  • Do some analysis about rare word embeddings analysis
  • Complete the code according to my analysis and experiments are running
  • Reproduce the paper's method and try to achieve some improvements if possible
Andi Zhang
Shiyao Li
2016/09/12 Yang Feng
  • surveyed several papers of NN parsing;
  • came up with the details of our memory network grammar
  • downloaded the resources of baselines, including: Dyer's rnn grammar, facebook memory network and facebook bAbI data
  • ran and read the code of rnn grammar
  • get the result of rnn grammar on bAbI data set;
  • understand the code of rnn grammar.
Jiyuan Zhang
  • wrote a report on the music generation [here]
  • completed the text section of my part of deep learning chapter,yet not completed the illustration [here]
  • complete the illustration
  • find some paper about generation task
Aodong Li
Andi Zhang
  • look for source code of Memnn
  • prepare for the sharing of a paper
Shiyao Li
Shiyue Zhang
  • Help Zhiyong with the document
  • Read papers: "Memory Networks" and "Neural Machine Translation"
  • Go on with the paper reading
  • prepare for the bi-week report