文章目錄

• • PAPER RECODED
  • • [NO.1] 【第一次找論文作者要源碼】
    • [NO.2] 【my blog】
    • [NO.3]【余弦相似度】
    • [NO.4]【Ablation Study】
    • [NO.5] How to wirte a Paper!
    • 【Part 2】 How to write a good review

PAPER RECODED

[NO.1] 【第一次找論文作者要源碼】

笑死自己了，在寫信的時候本來想開頭空格調整好，無意中多按了幾次回車，導致最終沒有調整好就把郵件發了出去！可能被大佬見笑了！還好是基本信息都有表達清楚！下次發重要郵件的時候一定不能把收件人的郵箱先填，不然無腦按觸發快捷鍵就悲劇了，現在就希望作者能回復我的消息le…

[NO.2] 【my blog】

個人博客網站:https://jucway.github.io/

[NO.3]【余弦相似度】

余弦相似度

[NO.4]【Ablation Study】

An ablation study typically refers to removing some “feature” of the model or algorithm, and seeing how that affects performance.

Examples:

An LSTM has 4 gates: feature, input, output, forget. We might ask: are all 4 necessary? What if I remove one? Indeed, lots of experimentation has gone into LSTM variants, the GRU being a notable example (which is simpler).
If certain tricks are used to get an algorithm to work, it’s useful to know whether the algorithm is robust to removing these tricks. For example, DeepMind’s original DQN paper reports using (1) only periodically updating the reference network and (2) using a replay buffer rather than updating online. It’s very useful for the research community to know that both these tricks are necessary, in order to build on top of these results.
If an algorithm is a modification of a previous work, and has multiple differences, researchers want to know what the key difference is.
Simpler is better (inductive prior towards simpler model classes). If you can get the same performance with two models, prefer the simpler one

[NO.5] How to wirte a Paper!

[Abrast]
[Introduction]
you must make your paper easy to read. You’ve

[Related work]
[Main idear]

show simple toy examples to let people get the main idear

[Image model]
[Algorithm]
[Experiments]
Gone are this days of ,“We think this is a great idea and we expect it will be very useful in computer vision. See how it works on this meaningless, contrived problem?”
[Discussion]
Future work?

[Conclsions]
Conclusions, or what this opens up, or how this can change how we appreach computer vision problems.

【How】

General writing tips
Keep the reader upper-most in your mind;

Treat the reader as you would a guest in your house;

盡量用簡短的單詞表達出你需要的東西就可以，下面就是比較這種方式；

精簡的語句表達出想要的結果

Glance at title
Skim abstract
Look at the figure
Read every word

the caption shoud tell the reader what to notice about the figure.

【Part 2】 How to write a good review

Summary Of the Contributions

• The Strengths
  Describe the strengths of the paper in detail

• The Weakness
  Describe the weaknesses of the paper in detail. Provide solid arguments and evidence for your claims.

  • It is not okay to say that something has been done before if you do not provide concrete references.

• 總結：

用自己的話把自己的貢獻描述清楚

如果是在之前的工作上做的改進，就可說明與之前的工作，這個工作有什么改變和改進，突出自己的差異和比較；

• 優勢： 把自己的優勢詳細的體現出來。
• 不足：

• Summarizing the paper

you should be able to descrite the paper in your own words

why is the paper intersting to the CVPR audient?

• Strengths:

the authors are trying real hard to tell you about these

• Weaknesses

Put on your most constructive "I’m an advisor and mentor"hat.

• Suggestion To Authors

Provide recommendations to the authors how they can improve their manuscript

Great opportunity to read other work
保持開放的心，閱讀其他reviews rebuttal, discuss the paper and update your review.

How to be a good reviewer

before during after three part to plan

A list of Questions

Is the problem clearly work cover the relevant approaches

How important is sata?

justification of design choices
contribution
self-make baselines


增量的實驗是不被允許的

novelty is in the eye of the beholder（創新在善于觀察的眼中）

Could this interest someone at CVPR!!!
是否審稿人會對這個感興趣
you rarely get chance to review such papers. Try not to mess it up!

Wrong
content organization not English

Helpful Guideline for Reviewing
-









sort by importance, and answer precisely concisely, unambiguously

總結

以上是生活随笔為你收集整理的[PAPER-RECORD]的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。