Videos

How Convolutional Neural Networks work



Brandon Rohrer

Find the rest of the How Neural Networks Work video series in this free online course:
https://end-to-end-machine-learning.teachable.com/p/how-deep-neural-networks-work

A gentle guided tour of Convolutional Neural Networks. Come lift the curtain and see how the magic is done. For slides and text, check out the accompanying blog post: http://brohrer.github.io/how_convolutional_neural_networks_work.html

Check out https://youtu.be/FmpDIaiMIeA for better audio and a more detailed account.

Follow me for announcements: https://twitter.com/_brohrer_

Source

Similar Posts

47 thoughts on “How Convolutional Neural Networks work
  1. You should also add the information about padding when doing the convolutions with the filters as we lose pixel information if we do not do any padding when we get to the edge pixels.

  2. at 16:00 why fully connected layers are stacked .one fully connected layer successfully decided that its a X by high average value .92. I m confuse about why we need again and again fully connected layers.

  3. I do not understand the voting part, how does it work and how backpropagation is used for getting votes values. I am pretty confused in it.
    Also, how the values of X or O is changing in the end-> for votes, when testing it??

  4. i don't know why there r lots of lots of knowledge about how to implement a cnn with lots of boring parameters we dont understand, no body tell us the reason behind it except this one .

  5. A true master are good at simplifying complex things . i have a question . how the computer choose the 2 diagonal line filters and middle little X filter since it have no eyes.

  6. Thank you so much!
    I have one question btw.
    So if the first conv layer had 5 features to check for and the second layer had 3 features, you would expect 15 outputs matrices right?

    Just wanted to make sure since with large numbers of features you can probably expect potentially several thousand matrices and that sounds insane, especially with multiple color channels.

  7. Incredibly interesting and well-made video. What I can't currently understand is how do multiple images(the training data) help with the back-propagation? Are the images looked at separately, with different weights adjusted for each image and then an average calculated, or another method? I'm thinking of making my own CNN, but I'm not exactly sure where the training data fits in. I would certainly appreciate any advice and/or explanations.

  8. Awesome video on convolution neural networks. Thanks so much Brandon, I was having difficulty to understand the layers but you made it crystal clear.

  9. I was fine up until 14:26 when you didn't explain how the voting for and X or an O works. Now I'm lost. Lines magically appear to the X and O without any reason as to why they're there.

  10. Ohmygoodgod, this video just saved a presentation of mine and made me feel like I can understand everything 😀 Thank you so much for this!!! Eternaly greatful!

  11. Great video. I feel like this is one of the most comprehensive especially on how it explains the process step by step (for instance i noticed other explanations don't bother with how stacking layers works)

  12. Thank you for the great lecture. At the end, you emphasized that CNN works for image (or data that can be expressed in image formats). For data that arrangement doesn't matter what methods do you suggest?

Comments are closed.

WP2Social Auto Publish Powered By : XYZScripts.com