Yiwei Lin
    • Create new note
    • Create a note from template
      • Sharing URL Link copied
      • /edit
      • View mode
        • Edit mode
        • View mode
        • Book mode
        • Slide mode
        Edit mode View mode Book mode Slide mode
      • Customize slides
      • Note Permission
      • Read
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
      • Write
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
    • Invite by email
      Invitee

      This note has no invitees

    • Publish Note

      Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

      Your note will be visible on your profile and discoverable by anyone.
      Your note is now live.
      This note is visible on your profile and discoverable online.
      Everyone on the web can find and read all notes of this public team.

      Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

      Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

      Explore these features while you wait
      Complete general settings
      Bookmark and like published notes
      Write a few more notes
      Complete general settings
      Write a few more notes
      See published notes
      Unpublish note
      Please check the box to agree to the Community Guidelines.
      View profile
    • Commenting
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
      • Everyone
    • Suggest edit
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
    • Emoji Reply
    • Enable
    • Versions and GitHub Sync
    • Note settings
    • Note Insights New
    • Make a copy
    • Transfer ownership
    • Delete this note
    • Save as template
    • Insert from template
    • Import from
      • Dropbox
      • Google Drive
      • Gist
      • Clipboard
    • Export to
      • Dropbox
      • Google Drive
      • Gist
    • Download
      • Markdown
      • HTML
      • Raw HTML
Menu Note settings Note Insights Versions and GitHub Sync Sharing URL Create Help
Create Create new note Create a note from template
Menu
Options
Make a copy Transfer ownership Delete this note
Import from
Dropbox Google Drive Gist Clipboard
Export to
Dropbox Google Drive Gist
Download
Markdown HTML Raw HTML
Back
Sharing URL Link copied
/edit
View mode
  • Edit mode
  • View mode
  • Book mode
  • Slide mode
Edit mode View mode Book mode Slide mode
Customize slides
Note Permission
Read
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
Write
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
  • Invite by email
    Invitee

    This note has no invitees

  • Publish Note

    Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

    Your note will be visible on your profile and discoverable by anyone.
    Your note is now live.
    This note is visible on your profile and discoverable online.
    Everyone on the web can find and read all notes of this public team.

    Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Explore these features while you wait
    Complete general settings
    Bookmark and like published notes
    Write a few more notes
    Complete general settings
    Write a few more notes
    See published notes
    Unpublish note
    Please check the box to agree to the Community Guidelines.
    View profile
    Engagement control
    Commenting
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    • Everyone
    Suggest edit
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    Emoji Reply
    Enable
    Import from Dropbox Google Drive Gist Clipboard
       Owned this note    Owned this note      
    Published Linked with GitHub
    • Any changes
      Be notified of any changes
    • Mention me
      Be notified of mention me
    • Unsubscribe
    --- tags: Deep learning --- # 論文閱讀: VGG(VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION) [論文](https://arxiv.org/pdf/1409.1556.pdf) ## Introduction VGGNet 承襲 AlexNet,提出了通過使用更小的kernel size(3x3的filter),使網路架構更有彈性,得以堆疊出深度更深,表現更好的CNN模型。 ## ConvNet Configurations ### Architecture * VGG的input為224 × 224的RGB image,在training set上會將原始pixel值減去平均的RGB值做pre-processing。 * 在VGG中,使用了大量的3x3 filter取代在AlexNet中較大(11x11、7x7、5x5)的convolution filter。使network在相同的receptive field下,可以通過多層非線性層,學習到更複雜的pattern。 * 此外也使用了1 × 1 的convolution filters,它可以被視為將input channel做linear transform,提高non-linearity。 * 下採樣的部分使用了Max-pooling(2 × 2 pixel window, stride 2) * VGG架構的最後三層為Fully-Connected layers: 其中前兩層各有4096個channels,最後一層則因應ILSVRC共1000種類別對應1000個channel,並且是一個soft-max層。 * 激活函數使用AlexNet中提到的ReLu。 * 論文中經過實驗認為: 在AlexNet中使用的LRN只會單純增加計算量,然而對於performance卻沒有幫助。 ### Configurations 如下表,論文中嘗試實驗不同深度的CNN。 至於每一層的寬度(width, the number of channels),從第一層的64 channel開始,每經過一層max-pooling就增加一倍,直到512 channel為止。 ![](https://i.imgur.com/F93x93Y.png) ### Discussion VGG中,最重要的想法就是使用大量的3 X 3 Conv。作者認為這些改變有幾個好處: 1. 我們可以利用小的Conv的組合,達到與大的conv相同的receptive field。如下圖,透過3x3的Conv組合,就足以表現出其他更大的Conv。並且這些組合的可能性超過直接使用大的Conv的數量。 ![](https://i.imgur.com/coO5YgX.png) 2. 使用較多的較小的Conv可以提高Non-Linearity。比起使用單個7x7的Conv,使用3個3x3的結合可以使decision function更discriminative(更有區別能力) 3. 比起使用相對大的Conv,使用小的Conv可以減少parameter。 * 假設input、output都是C個channel的話,使用7x7的單層Conv需要7^2^xC^2^ = 49xC^2^個parameters, * 然而若使用3個堆疊的3 × 3的話,parameters的數量為 3x(3^2^xC^2^) = 27xC^2^ 另外,論文通過1x1 conv的使用,在不會影響receptive的前提下,增加decision function的nonlinearity(P.S. 雖然論文中並非為此目的,但1x1 conv也可以被用來降維,減少network參數的使用,詳細可以閱讀[卷積神經網路(Convolutional neural network, CNN): 1×1卷積計算在做什麼](https://medium.com/@chih.sheng.huang821/%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E8%B7%AF-convolutional-neural-network-cnn-1-1%E5%8D%B7%E7%A9%8D%E8%A8%88%E7%AE%97%E5%9C%A8%E5%81%9A%E4%BB%80%E9%BA%BC-7d7ebfe34b8)。) ## Classification Framework (主要是如何調整參數,怎麼處理traning、testing data等細節,詳請閱讀論文) 稍微提及scale的方法: 由於network輸入必須為224x224,因此需對原始資料作調整。training方面,論文的做法是先將其scale到size S後,再把多出來的部分crop(類似AlexNet的做法)。並且有使用Multiple scale training: 在每次training 時,從一個固定的亂數範圍中,scale至範圍內的random值,再做crop。 然後輸入的test set則會scale到size Q,並使用多個crop進行預測,利用固定的crop大小預測左上、右上、左下、右下跟中間,並平均成最後預測結果。 training set的S與test set的Q不必相同。 ![](https://i.imgur.com/LQ2lEpU.png) ## Classification Experiment (此部分只大概整理一些重點,詳細請閱讀論文內容) ### Dataset 使用ILSVRC-2012 dataset。包含1000種類別資料,並且被分成training (1.3M images)、validation (50K images)、testing (100K images with held-out class labels)三種。 論文中絕大部分以validation set作為實驗的test set。 ### Single Scale Evaluation 首先,論文先使用了沒有scale的test set做實驗,結果如下表。(A-E對應上面架構的表格) ![](https://i.imgur.com/fQPzlfH.png) (表中S的[256;512]表示輸入的scale S會在兩數字間隨機選擇) 由於LRN的使用並沒有顯著的提升,因此作者在B-E的架構中就沒有實驗加入LRN的該種Network。 論文發現,雖著網路深度越深(從11層的A至19層的E),預測的結果也有所提升。 比較B與C的表現(有沒有1x1 Conv),我們可以看出1x1 Conv的加入對於表現有所提升。然而比較C(使用1x1 Conv)的表現卻比D(使用3x3 Conv)的差。論文由此推測,雖然額外的non-linearity有所幫助(C>B),但是萃取空間域的複雜特徵依然十分重要(D>C) ### Multi-Scale Evaluation 此部分中,test set中同一筆data會用不同的scale Q輸入,再把所有output(對於各個類別的score)平均起來作為最終結果。 結果如下圖,可以看到準確度有所提升。 ![](https://i.imgur.com/swL7jVd.png) ### Multi-Crop Evaluation 論文中也展示了input多種crop區域的Multi-Crop方法,以及結合Multi-Crop與Multi-Scale的結果,如下表。 ![](https://i.imgur.com/rGaopA8.png) ### ConvNet Fusion 如下表,論文也展示了使用多個model ensemble的結果。 ![](https://i.imgur.com/mmCmX8T.png) ### Comparison With The State Of Art 如下表,論文比較自己與當時的state-of-art的結果。 ![](https://i.imgur.com/O4hkhJC.png) ## Conclusion VGG展示了deeper network可以達到better result的效果。不過如果比較D跟E的結果,會發現performance的提升似乎已經到了緊繃。難道不是越深的網路越好嗎?這些問題在ResNet將得以解決。 另一方面,VGG透過使用大量3x3 filter替代size更大的filter,並證明其可以提升預測的表現,為deep learning模型的設計提出了重要的看法。 ## Reference [VGG_深度學習_原理](https://medium.com/@danjtchen/vgg-%E6%B7%B1%E5%BA%A6%E5%AD%B8%E7%BF%92-%E5%8E%9F%E7%90%86-d31d0aa13d88)

    Import from clipboard

    Paste your markdown or webpage here...

    Advanced permission required

    Your current role can only read. Ask the system administrator to acquire write and comment permission.

    This team is disabled

    Sorry, this team is disabled. You can't edit this note.

    This note is locked

    Sorry, only owner can edit this note.

    Reach the limit

    Sorry, you've reached the max length this note can be.
    Please reduce the content or divide it to more notes, thank you!

    Import from Gist

    Import from Snippet

    or

    Export to Snippet

    Are you sure?

    Do you really want to delete this note?
    All users will lose their connection.

    Create a note from template

    Create a note from template

    Oops...
    This template has been removed or transferred.
    Upgrade
    All
    • All
    • Team
    No template.

    Create a template

    Upgrade

    Delete template

    Do you really want to delete this template?
    Turn this template into a regular note and keep its content, versions, and comments.

    This page need refresh

    You have an incompatible client version.
    Refresh to update.
    New version available!
    See releases notes here
    Refresh to enjoy new features.
    Your user state has changed.
    Refresh to load new user state.

    Sign in

    Forgot password
    or
    Sign in via Google Sign in via Facebook Sign in via X(Twitter) Sign in via GitHub Sign in via Dropbox Sign in with Wallet
    Wallet ( )
    Connect another wallet

    New to HackMD? Sign up

    By signing in, you agree to our terms of service.

    Help

    • English
    • 中文
    • Français
    • Deutsch
    • 日本語
    • Español
    • Català
    • Ελληνικά
    • Português
    • italiano
    • Türkçe
    • Русский
    • Nederlands
    • hrvatski jezik
    • język polski
    • Українська
    • हिन्दी
    • svenska
    • Esperanto
    • dansk

    Documents

    Help & Tutorial

    How to use Book mode

    Slide Example

    API Docs

    Edit in VSCode

    Install browser extension

    Contacts

    Feedback

    Discord

    Send us email

    Resources

    Releases

    Pricing

    Blog

    Policy

    Terms

    Privacy

    Cheatsheet

    Syntax Example Reference
    # Header Header 基本排版
    - Unordered List
    • Unordered List
    1. Ordered List
    1. Ordered List
    - [ ] Todo List
    • Todo List
    > Blockquote
    Blockquote
    **Bold font** Bold font
    *Italics font* Italics font
    ~~Strikethrough~~ Strikethrough
    19^th^ 19th
    H~2~O H2O
    ++Inserted text++ Inserted text
    ==Marked text== Marked text
    [link text](https:// "title") Link
    ![image alt](https:// "title") Image
    `Code` Code 在筆記中貼入程式碼
    ```javascript
    var i = 0;
    ```
    var i = 0;
    :smile: :smile: Emoji list
    {%youtube youtube_id %} Externals
    $L^aT_eX$ LaTeX
    :::info
    This is a alert area.
    :::

    This is a alert area.

    Versions and GitHub Sync
    Get Full History Access

    • Edit version name
    • Delete

    revision author avatar     named on  

    More Less

    Note content is identical to the latest version.
    Compare
      Choose a version
      No search result
      Version not found
    Sign in to link this note to GitHub
    Learn more
    This note is not linked with GitHub
     

    Feedback

    Submission failed, please try again

    Thanks for your support.

    On a scale of 0-10, how likely is it that you would recommend HackMD to your friends, family or business associates?

    Please give us some advice and help us improve HackMD.

     

    Thanks for your feedback

    Remove version name

    Do you want to remove this version name and description?

    Transfer ownership

    Transfer to
      Warning: is a public team. If you transfer note to this team, everyone on the web can find and read this note.

        Link with GitHub

        Please authorize HackMD on GitHub
        • Please sign in to GitHub and install the HackMD app on your GitHub repo.
        • HackMD links with GitHub through a GitHub App. You can choose which repo to install our App.
        Learn more  Sign in to GitHub

        Push the note to GitHub Push to GitHub Pull a file from GitHub

          Authorize again
         

        Choose which file to push to

        Select repo
        Refresh Authorize more repos
        Select branch
        Select file
        Select branch
        Choose version(s) to push
        • Save a new version and push
        • Choose from existing versions
        Include title and tags
        Available push count

        Pull from GitHub

         
        File from GitHub
        File from HackMD

        GitHub Link Settings

        File linked

        Linked by
        File path
        Last synced branch
        Available push count

        Danger Zone

        Unlink
        You will no longer receive notification when GitHub file changes after unlink.

        Syncing

        Push failed

        Push successfully