Kolmogorov-Arnold Networks (KANs) - What are they and how do they work?

Поділитися
Вставка
  • Опубліковано 24 гру 2024

КОМЕНТАРІ • 56

  • @michaelzumpano7318
    @michaelzumpano7318 3 дні тому

    Your delivery was brilliant! You gave the right amount of detail. I’m really a fan now. Can’t wait for the next one, and I’m going to watch your other playlists. Thank you.

  • @AbdallahBoukouffallah
    @AbdallahBoukouffallah 21 день тому +3

    Truly a hidden gem. I've missed out a lot by not knowing about this channel, and off course Luis is the best math teacher i've ever had.

  • @brainxyz
    @brainxyz 21 день тому +1

    The best explanation of the topic on UA-cam. All you need to know is explained in less than 15 minute! Thank you very much

  • @brothberg
    @brothberg 3 дні тому

    Great delivery! I wish every math teacher was like Luis.

  • @SerranoAcademy
    @SerranoAcademy  12 днів тому

    The second part is out, on the Kolmogorov-Arnold Theorem!
    ua-cam.com/video/nS2hnm0JRBk/v-deo.htmlsi=ym6OsCVKFgiHhtne

  • @rikiakbar4025
    @rikiakbar4025 21 день тому

    I can't thank you enough, Luis. You make all this stuff look very simple.

  • @ufuoma833
    @ufuoma833 21 день тому +9

    and this lesson is free? What a time to be alive!

    • @SerranoAcademy
      @SerranoAcademy  21 день тому +3

      Thank you, my payment is kind comments like yours. :)

    • @ufuoma833
      @ufuoma833 21 день тому

      @@SerranoAcademy I'm actually looking for other architectures right now since my models can't get pass the 88% AUC ROC maximum. Hopefully, I can use this to get to that sweet 95-ish %. Thank you again. Please be kind with the maths in your next video. lol

  • @fangyuan871
    @fangyuan871 17 днів тому

    Thank you so much for your efforts to put out such informative videos.

  • @skydiver151
    @skydiver151 21 день тому +1

    Very well done video, as usual! Great and interesting work!

    • @SerranoAcademy
      @SerranoAcademy  21 день тому +2

      @@skydiver151 thank you! I’m glad you liked it!

  • @timothywcrane
    @timothywcrane 19 днів тому

    12/5. Checked the channel for #2. Eagerly waiting. Great Video.

  • @jbtechcon7434
    @jbtechcon7434 21 день тому +1

    excellent explanation. thank you so much

  • @frankl1
    @frankl1 21 день тому +2

    At first look, KAN requires more parameters to be trained than MLP, but they claimed in the paper that KAN can compete equally if not better with MLP using a smaller network, therefore a smaller number of layers. I cannot wait to watch the next video, I would like to understand how the initial splines are chosen. For instance, if we go with B-splines, which one do we take and how many ? Are there other parameters to learn in addition to the knots ?

  • @khaledal-utaibi2049
    @khaledal-utaibi2049 5 днів тому

    Thanks!

    • @SerranoAcademy
      @SerranoAcademy  5 днів тому

      @khaledal-utaibi2049 thank you so much for your very kind contribution! I really appreciate it ☺️

  • @figueraxiyana9411
    @figueraxiyana9411 16 днів тому

    Muchas gracias por las explicaciones. Como siempre, las mejores que hay en youtube.

    • @SerranoAcademy
      @SerranoAcademy  15 днів тому

      Muchas gracias, me alegra que te gusten! :)

  • @kanakraj3198
    @kanakraj3198 21 день тому

    This was a really good explanation of KANs. 🥳

  • @ምእንቲመጎጎትሕለፍኣንጭዋ

    Clearly explained and illustrated. Thank you.

  • @simoncaicedo3266
    @simoncaicedo3266 16 днів тому

    This is fantastic! Thank you

  • @ibrahimmosty1860
    @ibrahimmosty1860 21 день тому

    Amazing video by amazing teacher

    • @SerranoAcademy
      @SerranoAcademy  21 день тому +1

      Thank you! :) The next one is coming up soon, and I'm having a lot of fun making it. :)

  • @RB31557
    @RB31557 17 днів тому

    Sir: Around 11:28 in your video you show quadratic B splines (my question is true for any spline approximation) three splines that approximate the function of interest. I was unclear how this will be used. They will not be used as weights for a linear dot product right? The three splines connecting to x1 will be used to determine what each outputs right? If x1 cvalue is .3 then the middle will output a 0.3 and the other two will output 0. Am I right? I am confused how you can use them as weights in the regular sense.

  • @tantzer6113
    @tantzer6113 21 день тому

    Mathematical beauty is enough to motivate me to watch the rest of this series. But the practical question is whether these networks can perform as well as neural networks on benchmarks, given equal “compute.” That is probably more an empirical question than a mathematical one.

    • @SerranoAcademy
      @SerranoAcademy  21 день тому +1

      Thank you! I fully agree. I really liked the mathematical beauty, so that is what caught my interest. From what I understand, they perform well compared to regular NNs. But it could go either way; they could become huge, or not. However, my hope is that either way, they'll inspire new architectures coming from the theory of representation of functions, as this is a beautiful field that has remained (until now) unexplored in ML.

  • @snapshot8886
    @snapshot8886 12 днів тому

    Eagerly waiting for the second part!!!❤

    • @SerranoAcademy
      @SerranoAcademy  12 днів тому

      Great timing! The second part just came out! :) ua-cam.com/video/nS2hnm0JRBk/v-deo.html

    • @snapshot8886
      @snapshot8886 12 днів тому

      ​@@SerranoAcademy thank u sir,u r awesome!!!🎉

  • @keghnfeem4154
    @keghnfeem4154 16 днів тому

    Thanks prophet.

  • @khoakirokun217
    @khoakirokun217 21 день тому

    Why this video only have 813 views after 4 hours? Subscribe instantly :D

  • @kasraamanat5453
    @kasraamanat5453 21 день тому

    Thank you so much Luis , cant wait for next chapter 😍

    • @SerranoAcademy
      @SerranoAcademy  21 день тому +1

      Thank you! :) Yes, super excited for that one, it's coming up soon!

    • @kasraamanat5453
      @kasraamanat5453 17 днів тому

      @@SerranoAcademy 😍😍

  • @Sydra.
    @Sydra. 21 день тому +1

    I think they reinvented the wheel with this one. Existing NN-s are already KANs. What they think is new is a misunderstanding of these concepts.

  • @Pedritox0953
    @Pedritox0953 21 день тому

    Great video! Very well explained, Peace!

    • @SerranoAcademy
      @SerranoAcademy  21 день тому

      @@Pedritox0953 thank you so much, I’m glad you liked it! Peace! 😊

  • @robertholder
    @robertholder 20 днів тому

    Seems to me that, instead of training weights that lead to activation functions, KANs are training weights (knot vectors) that lead to spines. Interested to learn more about the tradeoffs between the two.

  • @neelkamal3357
    @neelkamal3357 20 днів тому

    thank you sir

  • @squarehead6c1
    @squarehead6c1 14 днів тому

    Typo in the KAN depiction: output should be f1(x1) + f2(x2).

    • @SerranoAcademy
      @SerranoAcademy  14 днів тому

      Oh thanks! Yeah you're right, the w's should be x's.

  • @DrBilalHussain
    @DrBilalHussain 21 день тому

    Amazing video! Thanks :)

  • @SriHariBH
    @SriHariBH 21 день тому

    You are the best!!!

  • @yashagrahari
    @yashagrahari 20 днів тому

    why do we require 4 basis functions to approximate any linear/quadratic function with 3 bins?

  • @GaneshMula2023
    @GaneshMula2023 21 день тому

    great video😊

  • @AJKirby64
    @AJKirby64 18 днів тому

    Ok this is great. However doesn't it also demonstrate that KANs and MLPs are equivalent? The spline sections are equivalent to the activation levels.. and the choice of b splines is equivalent to the choice of functions. So aren't the two theories and entire architectures potentially equivalent? Is this just a choice of how to get the same function approximation system into memory?

  • @xHubAI
    @xHubAI 21 день тому

    ❤️🧠

  • @carolinalasso
    @carolinalasso 24 дні тому

    🎉

  • @aminkazemi1803
    @aminkazemi1803 19 днів тому

    It is the same CNN+DENSE layers , so what is its advantage ? Would you please take example of its advantage?