Backpropagation in Convolutional Neural Networks (CNNs)

แชร์
ฝัง
  • เผยแพร่เมื่อ 26 พ.ย. 2024

ความคิดเห็น • 112

  • @louissimon2463
    @louissimon2463 ปีที่แล้ว +11

    great video, but i don't understand how we can find the value of the dL/dzi terms. At 7:20 you make it seem like dL/dzi = zi, is that correct?

    • @far1din
      @far1din  ปีที่แล้ว +5

      No, they come from the loss function. I explain this at 4:17.
      It might be a bit unclear so I’ll highly reccomend you watch the video from 3blue1brown: th-cam.com/video/tIeHLnjs5U8/w-d-xo.htmlsi=Z6asTm87XWcW1bVn 😃

    • @rtpubtube
      @rtpubtube 10 หลายเดือนก่อน +4

      I'm with @louissimion, you show how dL/dw1 is related to dz1/dw1+... (etc), but you never show/expain where dL/dz1 (etc) comes from. Poof - miracle occurs here. Having a numerical example would help a lot. This "theory/symbology" only post is therefore incomplete/useless from a learing/understanding standpoint.

    • @mandy11254
      @mandy11254 6 หลายเดือนก่อน +2

      ​@@rtpubtubeIt's quite literally what he wrote. He hasn't defined a loss function so that's just what it is from the chain rule. If you're asking how the actual value of dL/dz1 is computed, the last layer has its own set of weights besides the ones shown in the video, in addition to an activation function. You use that and a defined loss function to compute dL/dzi. It's similar to what you see in standard NNs. If you studied neural networks, you should know this. This is a video about CNNs not an intro to NNs. Go study that before this. It's not his job to point out every little thing.

  • @khayyamnaeem5601
    @khayyamnaeem5601 ปีที่แล้ว +15

    Why is this channel so underrated? You deserve more subscribers and views.

    • @eneadriancatalin
      @eneadriancatalin ปีที่แล้ว +1

      Perhaps developers use ad blockers, and as a result, TH-cam needs to ensure revenue by not promoting these types of videos (that's my opinion)

  • @nizamuddinkhan9443
    @nizamuddinkhan9443 ปีที่แล้ว +4

    Very well explanation, I search many videos but no body explained regarding change in filter's weight. Thank you so much for this animated simple explanation.

  • @abhimanyugupta532
    @abhimanyugupta532 6 หลายเดือนก่อน +2

    Been trying to understand backpropogation in CNN for years until today! Thanks a ton mate!

    • @yosukesharp
      @yosukesharp 6 หลายเดือนก่อน

      it was obvious primitive algo dude... people like you are being called "data scientists" now, which is really sad...

    • @blubaylon
      @blubaylon หลายเดือนก่อน

      ​@@yosukesharpget off your high horse

  • @JessieJussMessy
    @JessieJussMessy ปีที่แล้ว +2

    This channel is a hidden gem. Thank you for your content

  • @haideralix
    @haideralix ปีที่แล้ว +2

    I have seen few videos before, this one is by far the best one. It breaks down each concept and answers all the questions that comes in the mind. The progression, the explanation is best

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you! 🔥

  • @srinathchembolu7691
    @srinathchembolu7691 3 หลายเดือนก่อน

    This is gold. Watching this after reading Michael Nielsen makes the concept crystal clear

  • @zemariamm
    @zemariamm ปีที่แล้ว +3

    Fantastic explanation!! Very clear and detailed, thumbs up!

  • @Joker-ez2fm
    @Joker-ez2fm 11 หลายเดือนก่อน

    Please do not stop making these videos!!!

    • @far1din
      @far1din  11 หลายเดือนก่อน +1

      I won’t let you down Joker 🔥🤝

  • @noohayub2188
    @noohayub2188 หลายเดือนก่อน

    what an excellent demonstration of the backpropagation on CNN, you won my heart, literally no one on the internet explains it as clearly as you did but please try to make another video as a sequel to this one where you also use the biases, a more complex exampl

  • @boramin3077
    @boramin3077 4 หลายเดือนก่อน

    Best video to understand what is going on the under the hood of CNN.

  • @heyman620
    @heyman620 ปีที่แล้ว +2

    What a masterpiece.

  • @jayeshkurdekar126
    @jayeshkurdekar126 ปีที่แล้ว +2

    You are a great example of fluidity of thought and words..great explanation

    • @far1din
      @far1din  ปีที่แล้ว +1

      Thank you my friend. Hope you got some value! :)

    • @jayeshkurdekar126
      @jayeshkurdekar126 ปีที่แล้ว

      @@far1din sure did

  • @giacomorotta6356
    @giacomorotta6356 ปีที่แล้ว

    great video, underrated channel , please keep it up with CNN videos!

  • @markuskofler2553
    @markuskofler2553 ปีที่แล้ว +5

    Couldn’t explain it better myself … absolutely amazing and comprehensible presentation!

  • @Peterpeter-hr8gg
    @Peterpeter-hr8gg ปีที่แล้ว +1

    what i was looking for. well explained

  • @PlabonTheSadEngineer
    @PlabonTheSadEngineer 10 หลายเดือนก่อน +1

    please continue your videos !!

  • @rubytejackson
    @rubytejackson 2 หลายเดือนก่อน

    This is an exceptional explanation, and I can't thank u more... u have to keep going, u enlighten many student on the planet! that's the best thing a human can do!

    • @far1din
      @far1din  2 หลายเดือนก่อน

      Thank you brother, very much appreciate it! 🔥

  • @shivanisrivarshini180
    @shivanisrivarshini180 29 วันที่ผ่านมา

    Great explanation. Thank you sir

  • @saikoushik4064
    @saikoushik4064 9 หลายเดือนก่อน

    Great Explanation, helped me understand the background working

  • @paedrufernando2351
    @paedrufernando2351 ปีที่แล้ว

    your channel is a Hidden Gem..My suggestion is to start a discord and get some crowd functing and one on ones for people who want to learn from you..youa re gifted in teaching.

  • @SolathPrime
    @SolathPrime ปีที่แล้ว +7

    Well explained now I need to code it my self

    • @far1din
      @far1din  ปีที่แล้ว +4

      Haha, that’s the hard part

    • @SolathPrime
      @SolathPrime ปีที่แล้ว +6

      @@far1din I think I came up with a solution Here
      def backward(self, output_gradient, learning_rate):
      kernels_gradient = np.zeros(self.kernels_shape)
      input_gradient = np.zeros(self.input_shape)
      for i in range(self.depth):
      for j in range(self.input_depth):
      kernels_gradient[i, j] = convolve2d(self.input[j], output_gradient[i], "valid")
      input_gradient[j] += convolve2d(output_gradient[i], self.kernels[i, j], "same")
      self.kernels -= learning_rate * kernels_gradient
      self.biases -= learning_rate * output_gradient
      return input_gradient
      First i initialized the kernel gradient as an array of zeros with the kernel shape
      then I iterated through the depth of the kernels the the depth of the input then for each gradient withe respect to the kernel
      I did the same to compute the input gradients
      Your vid helped me understand the backward method better
      So I have to say thank you sooo much for it

    • @SolathPrime
      @SolathPrime ปีที่แล้ว

      @@far1din I'll document the solution and but it here when I do please pin the comment

    • @far1din
      @far1din  ปีที่แล้ว +1

      @@SolathPrime That’s great my friend. Will pin 💯

  • @DSLDataScienceLearn
    @DSLDataScienceLearn 10 หลายเดือนก่อน

    great explanation, clear direct and understandable, sub!

  • @sourabhverma9034
    @sourabhverma9034 6 หลายเดือนก่อน

    Really intuitive and great animations.

  • @farrugiamarc0
    @farrugiamarc0 8 หลายเดือนก่อน +1

    This is a topic which is rarely explained online, but it was very clearly explained here. Well done.

  • @shazzadhasan4067
    @shazzadhasan4067 ปีที่แล้ว

    Great explanation with cool visual. Thanks a lot.

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you my friend 😃

  • @DVSS77
    @DVSS77 ปีที่แล้ว

    really clear explanation and good pacing. I felt I understood the math behind back propagation for the first time after watching this video!

  • @ramazanyel5979
    @ramazanyel5979 6 หลายเดือนก่อน

    excellent. the exact video i was looking for.

  • @aikenkazin4096
    @aikenkazin4096 ปีที่แล้ว

    Great explanation and visualization

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you my friend 🔥🚀

  • @pedroviniciuspereirajunho7244
    @pedroviniciuspereirajunho7244 11 หลายเดือนก่อน

    Amazing!
    I was looking for some material like this a long time ago and only found it here, beautiful :D

    • @far1din
      @far1din  11 หลายเดือนก่อน +1

      Thank you my brother 🔥

  • @RAHUL1181995
    @RAHUL1181995 ปีที่แล้ว

    This was really helpful....Thank you so much for the vizualization...Keep up the good work...Looking forward to your future uploads.

  • @gregorioosorio16687
    @gregorioosorio16687 ปีที่แล้ว +1

    Thanks for sharing!

  • @bamboooooooooooo
    @bamboooooooooooo 7 หลายเดือนก่อน

    great job. this explanation is really intuitive

  • @chatgpt-nv5ck
    @chatgpt-nv5ck หลายเดือนก่อน

    Beautiful🙌

    • @far1din
      @far1din  หลายเดือนก่อน

      Thank you 🙌

  • @guoguowg1443
    @guoguowg1443 7 หลายเดือนก่อน

    great stuff man, crystal clear!

  • @AlbertoOrtiz-we2jc
    @AlbertoOrtiz-we2jc หลายเดือนก่อน

    excellent explanation thanks

    • @far1din
      @far1din  หลายเดือนก่อน

      Glad it was helpful!

  • @elgs1980
    @elgs1980 ปีที่แล้ว +1

    Thank you so much!!! This video is so so so well done!

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you. Hope you got some value out of this! 💯

  • @MarcosDanteGellar
    @MarcosDanteGellar ปีที่แล้ว

    the animations were super useful, thanks!

  • @mahmoudhassayoun9475
    @mahmoudhassayoun9475 2 หลายเดือนก่อน

    Good job , the explanation is super, I hope you do not stop making videos in this calibre . Did you use manim to make this video or an other video editor?

  • @LeoMarchyok-od5by
    @LeoMarchyok-od5by 7 หลายเดือนก่อน

    Best explanation

  • @aliewayz
    @aliewayz 6 หลายเดือนก่อน

    really beautiful, thanks.

  • @bug8628
    @bug8628 หลายเดือนก่อน

    Amazing video!! :D

    • @far1din
      @far1din  หลายเดือนก่อน

      Thanks! 😄

  • @osamamohamedos2033
    @osamamohamedos2033 7 หลายเดือนก่อน

    Masterpiece 💕💕

  • @AsilKhalifa
    @AsilKhalifa 4 หลายเดือนก่อน

    Thanks a lot!

  • @UtkalSinha-s8j
    @UtkalSinha-s8j ปีที่แล้ว

    Nicely put, thank you so much.

  • @objectobjectobject4707
    @objectobjectobject4707 ปีที่แล้ว

    Great example thanks a lot

  • @harshitbhandi5005
    @harshitbhandi5005 ปีที่แล้ว

    great explanation

  • @manfredbogner9799
    @manfredbogner9799 หลายเดือนก่อน

    Sehr gut

  • @rodrigoroman4886
    @rodrigoroman4886 ปีที่แล้ว +2

    Great video!! Your explanation is the best I have found.
    Could you please tell me what software you use for the animations ?

    • @far1din
      @far1din  ปีที่แล้ว

      I use manim 😃
      www.manim.community

  • @akshchaudhary5444
    @akshchaudhary5444 10 หลายเดือนก่อน

    amazing video thanks!

  • @dhudach
    @dhudach 2 หลายเดือนก่อน

    I'm new to machine learning and neural networks. Your video is very helpful. I have built a small python script just using numpy and I can train numerous samples. So this is a big picture question. Let's say I've trained my program on thousands of inputs and I'm satisfied. Now I want to see if it can recognize a new input, one not used in training. What weight and bias values do I use? After I'm finished with training, how do I modify the script to 'guess?' It would seem to me that back propagation isn't used because I don't actually have a 'desired' value so I'm not going to calculate loss. What weight and bias values do I use from the training sessions? There are dozens of videos and tutorials on training but I think the missing piece is what to do with the training program to make it become the 'trained' program, the one that guesses new inputs without back propagation.

  • @samiswilf
    @samiswilf ปีที่แล้ว

    Well done.

  • @ManishKumar-pb9gu
    @ManishKumar-pb9gu 10 หลายเดือนก่อน

    thanku you so much for this

  • @ziligao7594
    @ziligao7594 6 หลายเดือนก่อน

    Amazing

  • @yuqianglin4514
    @yuqianglin4514 ปีที่แล้ว

    fab video! help me a lot

    • @far1din
      @far1din  ปีที่แล้ว

      Glad to hear that you got some value out of this video! :D

  • @MohamedBENELMOSTAPHA-l4v
    @MohamedBENELMOSTAPHA-l4v 8 หลายเดือนก่อน

    I've had no trouble learning about the 'vanilla' neural networks. Although your videos are great, I can't seem to find resources that delve a little deeper into the explanations of how CNNs work. Are there any resources you would recommend ?

  • @r0cketRacoon
    @r0cketRacoon 4 หลายเดือนก่อน

    tks u very much for this video, but it's probably more helpful if you also add a max pooling layer.

  • @im-Anarchy
    @im-Anarchy ปีที่แล้ว +2

    perfect, one suggestion make videos a little longer 20-30 is a good number

    • @far1din
      @far1din  ปีที่แล้ว +1

      Haha, most people don't like these kind of videos too long. Average watchtime for this video is about 3minutes :P

    • @im-Anarchy
      @im-Anarchy ปีที่แล้ว +1

      ​@@far1din​oh shii! 3 minutes, that was very unexpected, maybe it's because people revisit the video to revise specific topic.

    • @far1din
      @far1din  ปีที่แล้ว

      Must be 💯

  • @arektllama3767
    @arektllama3767 ปีที่แล้ว +2

    1:15 why do you iterate in steps of 2? If you iterated by 1 then you could generate a 3x3 layer image. Is that just to save on computation time/complexity or is there something other reason for it?

    • @far1din
      @far1din  ปีที่แล้ว +3

      The reason why I used a stride of two (iterations in steps of two) in this video is partially random and partially because I wanted to highlight that the stride when performing backpropagation should be the same as when performing the forward propagation. In most learning materials I have seen, they usually use a stride of one, hence a stride of one for the backpropagation. This could lead to confusion when operating with larger strides.
      The stride could technically be whatever you like (as long as you keep it within the dimensions of the image/matrix). I could have chosen another number for the stride as you suggested. In that case, with a stride of one, the output would be a 3 x 3 matrix/image. Some will say that a shorter stride will encapsulate more information than a larger one, but this becomes “less true” as the size of the kernel increases. As far as I know there are no “rules” for when to use larger strides and not. Please let me know if this notion has changed as everything changes so quickly in this field! 🙂

    • @arektllama3767
      @arektllama3767 ปีที่แล้ว +3

      @@far1din I never considered how stride length could change depending on kernel size. I guess that makes sense, the larger kernel could cover the same data as a small kernel, just in fewer steps/iterations. I also figured you intentionally generated a 2x2 image since that’s a lot simpler than a 3x3 and this an educational video. Thanks for the feedback, that was really insightful!

  • @piyushkumar-wg8cv
    @piyushkumar-wg8cv ปีที่แล้ว +1

    Great explanation. Can you please tell which tool do you use for making these videos.

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you my friend! I use manim 😃
      www.manim.community

  • @OmidDavoudnia
    @OmidDavoudnia 7 หลายเดือนก่อน

    Thanks.

  • @vishvadoshi976
    @vishvadoshi976 2 หลายเดือนก่อน

    “Beautiful, isn’t it?”

  • @govindnair5407
    @govindnair5407 8 หลายเดือนก่อน

    What is the loss function here, and how are the values in the flattened z matrix used to compute yhat ?

  • @SiddhantSharma181
    @SiddhantSharma181 5 หลายเดือนก่อน

    Is the stride only along the rows, and not along columns? Is is common or just simplified?

  • @simbol5638
    @simbol5638 11 หลายเดือนก่อน

    +1 sub, excellent video

    • @far1din
      @far1din  11 หลายเดือนก่อน

      Thank you! 😃

  • @ItIsJan
    @ItIsJan ปีที่แล้ว

    5:24
    does this just mean we divide z1 by w1 and ultiply by L divided by z1 and do that for all z'S to get the partial derivative of L in respect to w1?

    • @far1din
      @far1din  ปีที่แล้ว

      It’s not that simple. Doing the actual calculations is a bit more tricky. Given no activation function, Z1 = w1*pixel1 + w2*pixel2 + w3*pixel3… you now have to take the derivative of this with respect to w1, then y = z1*w21 + z2*w22… take the derivative of y with respect to z1 etc. The calculus can be a bit too heavy for a comment like this.
      I’ll highly reccomend you watch the video by 3blue1brown: th-cam.com/video/tIeHLnjs5U8/w-d-xo.htmlsi=Z6asTm87XWcW1bVn 😃

  • @PeakyBlinder-lz2gh
    @PeakyBlinder-lz2gh 10 หลายเดือนก่อน

    thx

  • @bnnbrabnn9142
    @bnnbrabnn9142 8 หลายเดือนก่อน

    What about the weights of the fully connected layer

    • @mandy11254
      @mandy11254 6 หลายเดือนก่อน

      No point in adding it to this video since that's something you should know from neural networks. That's why he just leaves it as dL/dzi.

  • @MoeQ_
    @MoeQ_ ปีที่แล้ว +2

    dL/dzi = ??

    • @far1din
      @far1din  ปีที่แล้ว +1

      I explain the term at 4:17.
      It might be a bit unclear so I’ll highly reccomend you watch the video from 3blue1brown: th-cam.com/video/tIeHLnjs5U8/w-d-xo.htmlsi=Z6asTm87XWcW1bVn 😃

  • @burerabiya7866
    @burerabiya7866 ปีที่แล้ว

    Hello well explained. I need your presentation

    • @far1din
      @far1din  ปีที่แล้ว

      Just download it 😂

  • @Тима-щ2ю
    @Тима-щ2ю ปีที่แล้ว

    You have nices videos, that helped me better understand the concept of CNN. But, from this video, it is not really obvious that matrix dL/dw - is convolution of image matrix and dL/dz matrix, as showed here th-cam.com/video/Pn7RK7tofPg/w-d-xo.html. The stride of two is also a little bit confusing

    • @far1din
      @far1din  ปีที่แล้ว

      Thank you for the comment! I believe he is doing the exact same thing (?)
      I chose to have a stride of two in order to highlight that the stride should be similar to the stride used during the forward propagation. Most examples stick with a stride of one. I now realize it might have caused some confusion :p

  • @minhnguyenvu9479
    @minhnguyenvu9479 19 วันที่ผ่านมา

    original is a matrix of 5x5, kernel is a matrix of 3x3, then output must be a matrix of (5-3+1) x (5-3+1) or 3x3, not 2x2 as your video

    • @far1din
      @far1din  18 วันที่ผ่านมา

      The stride used on the example in this video is 2, hence the 2x2 output. You would have been correct if the stride was 1 😄

  • @CorruptMem
    @CorruptMem ปีที่แล้ว

    I think it's spelled "Convolution"

    • @far1din
      @far1din  ปีที่แล้ว +1

      Haha thank you! 🚀

  • @int16_t
    @int16_t ปีที่แล้ว

    w^* is an abuse of math notation, but it's convenient.