Optimizers - EXPLAINED!

Поділитися
Вставка
  • Опубліковано 19 лис 2024

КОМЕНТАРІ • 138

  • @varshitakolipaka7933
    @varshitakolipaka7933 29 днів тому +3

    This is THE best video I’ve watched on this topic: clear, perfectly motivated, and insanely engaging

  • @tombratfred3102
    @tombratfred3102 4 роки тому +142

    I like how you anthropomorphize optimizers. Makes me really empathize with their struggles.

    • @erich_l4644
      @erich_l4644 4 роки тому +6

      with a profile pic like that- you would

    • @metaprog46and2
      @metaprog46and2 4 роки тому +5

      @@erich_l4644 LMAO - your comment just won the internet. You'll soon receive an email by a Nigerian Prince with instructions on how to claim your winnings lol.

    • @metaprog46and2
      @metaprog46and2 4 роки тому +2

      Died laughing like thrice. Witty joke.

  • @jonass1315
    @jonass1315 2 роки тому +15

    This is how every lecture should be like. Funny but perfectly explained, and greatly visualized. Thanks!

  • @diaojun161
    @diaojun161 4 роки тому +51

    The best explaination of optimizers in DL I HAVE EVER WATCHED!

    • @CodeEmporium
      @CodeEmporium  4 роки тому +2

      Thank you! More of this to come!

  • @1harru
    @1harru 3 роки тому +10

    Hands down.. This is the best video on Optimizers.. !!! I've been trying to understand the complex math equations for the past few days and this one literally gave me the overall intuition in 7min 🙏🙏🙏

  • @dude8309
    @dude8309 4 роки тому +26

    great little overview! love how you get to the point quickly yet provide all the needed intuition

    • @CodeEmporium
      @CodeEmporium  4 роки тому +4

      Thanks! That's exactly what I was going for :)

  • @ZaidKhan-0407
    @ZaidKhan-0407 8 днів тому +1

    NGL this is one of the best video for Optimizers

  • @kevinelkin3943
    @kevinelkin3943 4 роки тому +7

    Such an underrated channel! Great explanations and visuals!

  • @zhengyahnis848
    @zhengyahnis848 3 роки тому +1

    Don't know why this video is under spreading, the explanation is great and the high-level summarization helps me a lot.

    • @CodeEmporium
      @CodeEmporium  3 роки тому

      Thank you! Mind fixing that by sharing this around? Would love to get more eyeballs here :)

  • @X_platform
    @X_platform 4 роки тому +25

    Loving the sound effect

  • @carebox6187
    @carebox6187 4 роки тому +16

    This video was both informative and hilarious. I absolutely loved it!

    • @CodeEmporium
      @CodeEmporium  4 роки тому +1

      That was the objective. Glad you liked it :)

  • @abhikbanerjee3719
    @abhikbanerjee3719 4 роки тому +6

    I am watching this at 2 am in the morning and that sudden effect 00:13 cracked me up!

  • @fahdciwan8709
    @fahdciwan8709 4 роки тому +1

    thanks! one of those rare videos that explain the intuition perfectly instead of hovering around the terms

  • @shubhigautam9655
    @shubhigautam9655 3 роки тому +1

    the only video thats ever made me laugh while explaining a concept. Love it, thank you!

  • @a.h.s.3006
    @a.h.s.3006 3 роки тому +2

    That...... was........ one EXCELLENT VIDEO!!!!!
    Thank you so much, I thought I would struggle with optimizers but now it's all clear to me

  • @carlavirhuez4785
    @carlavirhuez4785 4 роки тому +1

    Best video ever on optimizer. Thanks a lot.

  • @trocketflicks
    @trocketflicks 4 роки тому +1

    Man, this video is slept on. Such a good explanation!

  • @lakshmisrinivas369
    @lakshmisrinivas369 3 роки тому +3

    Absolute way of learning with lot of fun. Thanks for such a funny and insightful video

  • @BlockDesignz
    @BlockDesignz 4 роки тому +4

    Absolutely love this iterative explanation.

    • @CodeEmporium
      @CodeEmporium  3 роки тому +2

      Thank you. I'm experimenting with different teaching styles :)

  • @rendevous9253
    @rendevous9253 3 роки тому

    Man you gave the best explanation which even a noob like me in machine learning can understand . Keep it up man 👍.

  • @DouweMr
    @DouweMr 2 роки тому

    This is one hell of a video to refresh on this stuff! kindly appreciated!!

  • @Anja5233
    @Anja5233 Рік тому

    This is my new favorite video on the internet

    • @CodeEmporium
      @CodeEmporium  Рік тому

      Thanks so much for the compliments:) I try

  • @ruxiz2007
    @ruxiz2007 4 роки тому

    This video is so good, and it deserves 100X more attention!

  • @10bokaj
    @10bokaj 3 роки тому +1

    Very clear, very well explained 10/10

  • @shouravpaul3092
    @shouravpaul3092 4 роки тому +1

    like your video, and mostly i wanted to see the graph that most people dont show, thank you

  • @ahmedaj2000
    @ahmedaj2000 Рік тому

    love it! thank you! explained better than my professors. i finally get these now after so long

    • @CodeEmporium
      @CodeEmporium  Рік тому

      Words that are too kind. Thank for the kind words

  • @anujlahoty8022
    @anujlahoty8022 2 роки тому +1

    Very well explained and in a fun way.

  • @ErturkKadir
    @ErturkKadir 4 роки тому

    Such a clear and simple explanation of complicated things. Great job.

  • @igorg4129
    @igorg4129 Рік тому +2

    I think that a critical point missing here in the explanation:
    You have forgotten to mention that a loss surface is different for each sample so there DOES NOT EXIST any universal loss surface for a given dataset and this is a problem in stochastic gradient descent

  • @GauravSharma-ui4yd
    @GauravSharma-ui4yd 4 роки тому +2

    Thanks ajay for giving this a shoot. Loved it❤️

    • @CodeEmporium
      @CodeEmporium  4 роки тому +1

      Thanks for watching Gaurav (and the suggestion). Saw your comment on the last video too. And it was also in a line of videos I wanted to do. Probably not as "mathematical" as you'd like. I wanted to just explain why certain terms appear the way they do. Hopefully this helped that understanding. I might do a more mathematical video in the future though. But for now, this will do :)

    • @GauravSharma-ui4yd
      @GauravSharma-ui4yd 4 роки тому

      @@CodeEmporium You did a pretty awesome job in just 7 minutes. Its both beginner friendly and refreshing for intermediates.

  • @FanOfFunBuddy
    @FanOfFunBuddy 2 роки тому

    Wow most complex topic in under 7 minutes 😊 with pretty good visualizations.

  • @ZobeirRaisi
    @ZobeirRaisi 4 роки тому

    Your Explanation went to the deep of my brain!

  • @Hariharan-yy1fu
    @Hariharan-yy1fu 2 роки тому

    Awesome work easy to get a quick review before my interview keep going

  • @inteligenciamilgrau
    @inteligenciamilgrau 2 роки тому

    Best explanation ever!! Thank you so much!!!

  • @oskarbartosz9159
    @oskarbartosz9159 Рік тому

    m8, i was searching for channel like that for a really long time

  • @sb7048
    @sb7048 4 роки тому +1

    What does the alpha at the SGD momentum equation do? I mean alpha is your learning rat at the first two equations but since then you use n as the learning rate, so for what is alpha since then?

  • @eyesyt7571
    @eyesyt7571 Рік тому

    The first scene is precisely what happened to my neural network 2 weeks ago.

  • @TawhidShahrior
    @TawhidShahrior 2 роки тому +1

    man you do great work!

  • @crashedbboy
    @crashedbboy 10 місяців тому

    Never thought I would spit out drink while watching machine learning video

  • @MuhammadMujahidHaruna
    @MuhammadMujahidHaruna 2 місяці тому

    😮 Am silence of how you explain everything in details

  • @sia7001
    @sia7001 4 роки тому +2

    So underrated 😭

  • @m.a.flores7252
    @m.a.flores7252 4 роки тому

    Please keep doing this kind of videos I’m in love with ML and with u

    • @CodeEmporium
      @CodeEmporium  4 роки тому

      Haha thank you so much for the support

  • @Fransphoenix
    @Fransphoenix Рік тому

    Great explanation and fun, too. Thank you!

  • @r.y.y8073
    @r.y.y8073 Рік тому

    I like how you explained this!

  • @eniolaajiboye4399
    @eniolaajiboye4399 3 роки тому

    ❤️ the videos man. They're so clear

  • @hariharans.j5246
    @hariharans.j5246 4 роки тому +1

    do Neural ODEs and self-supervised learning techniques pls,
    great video btw

    • @CodeEmporium
      @CodeEmporium  4 роки тому +2

      Thanks. I saw your comment on another video. I'll look into this a bit

  • @mennoliefstingh5687
    @mennoliefstingh5687 4 роки тому +1

    Thanks for the great explanation!

  • @SujayAmberkar
    @SujayAmberkar 4 роки тому +2

    you voice is like some cool anime main character. I wished i had a voice like yours. Anyway great explanation.

  • @aafaq97in
    @aafaq97in 4 роки тому +1

    amazing vid you just earned a subscriber! looking forward to more content like this!

  • @roshanid6523
    @roshanid6523 3 роки тому

    Amazing explanation

  • @ThamizhanDaa1
    @ThamizhanDaa1 3 роки тому +1

    Nice channel! better than my professors lol

    • @CodeEmporium
      @CodeEmporium  3 роки тому +1

      Super happy this is helpful. Thanks!

  • @YangQuanChen
    @YangQuanChen 3 роки тому

    Nicely done! Thanks!

  • @rakeshsinghrawat99
    @rakeshsinghrawat99 4 роки тому +2

    Always good

  • @Simon-ed6zc
    @Simon-ed6zc 3 роки тому +2

    Hey, thank you a lot for the explanations! Do you happen to know any heuristics with which to chose a specific optimizer? Right now I have a problem where every paper uses Natural Gradient descent, but when I use it it barely ever converges, while Adam always gets it right (or at least comes close)...

    • @SirPlotsalot
      @SirPlotsalot 2 роки тому

      Your implementation might not be ideal, I'd try to use a KFAC preconditioning term maybe?

  • @dan1ar
    @dan1ar Рік тому

    why gradient on 1:21 would be large? Isn't it just avg for every element in dataset? Same for mini-batch, but only there gradient is avg for every element in batch

  • @ardhidattatreyavarma5337
    @ardhidattatreyavarma5337 Рік тому

    awesome explanation

  • @mahdijavadi2747
    @mahdijavadi2747 4 роки тому

    Thanks for the great clarification!

  • @rutweeksawant6567
    @rutweeksawant6567 4 роки тому

    very nice explanation and visualization.

  • @zshahlaie4740
    @zshahlaie4740 Рік тому

    this video was the bestttttt

  • @MrSinalta
    @MrSinalta 8 місяців тому

    If I understood well, Acceleration should not be called 'deceleration' in this particular case ?

  • @mridulavijendran3062
    @mridulavijendran3062 4 роки тому +2

    Hey. Great work on the video :D It was v clear and fascinating
    What's NAG? I wonder how come Nadam isn't popular -seems like a better choice.
    How would you describe RMSProp? You seem to have really great insight into DL concepts :D
    Also why expectation in particular for Adam parameter updates?
    Sorry for the questions bombardment. Jus pretty curious

  • @kidsfree6615
    @kidsfree6615 Рік тому

    Amazing video.

  • @lucha6262
    @lucha6262 4 роки тому

    really good overview

    • @CodeEmporium
      @CodeEmporium  4 роки тому

      Thanks! Making more of this stuff on the channel

  • @user-wr4yl7tx3w
    @user-wr4yl7tx3w 2 роки тому

    Well explained.

  • @karamjeetsinghgulati6989
    @karamjeetsinghgulati6989 2 роки тому

    I can start laughing for intial 20 sec, I am watching in loop

  • @chinmayeedongre5525
    @chinmayeedongre5525 4 роки тому

    Great Video!

  • @youtubecommenter5122
    @youtubecommenter5122 4 роки тому

    What a good video!

  • @nikab1852
    @nikab1852 3 роки тому

    love this

  • @ArsalJalib
    @ArsalJalib 4 роки тому

    Loved the start, watched 5 times.
    Also my first comment on UA-cam. =)

    • @CodeEmporium
      @CodeEmporium  3 роки тому

      Yas! Thank for this comment! Absolutely love it

  • @sizhuanghe1361
    @sizhuanghe1361 Рік тому

    That's fun!

  • @shaflyhamzah3848
    @shaflyhamzah3848 4 роки тому

    Nice explanation

  • @davisburnside9609
    @davisburnside9609 3 роки тому

    very helpful, thank you

  • @MrAdhito
    @MrAdhito 3 роки тому

    0:43 this really cracks me up HAHA!

  • @negardeylami6039
    @negardeylami6039 2 роки тому

    LMFAO the ontro I don't know why it's so funnyyyyyy for me

  • @RohitashChandra
    @RohitashChandra 4 роки тому

    well done!

  • @sumod12
    @sumod12 3 роки тому

    Awesome ❣️❣️

  • @SeanKearney-g7d
    @SeanKearney-g7d Місяць тому

    excellent

  • @yahavx
    @yahavx Рік тому

    The first part is not correct: the fact that you use a mini-batch in each step, rather than the entire dataset, does not give you a higher chance to converge to the optimum. Because even when considering the entire dataset in each step, you're still taking the average gradient, so the expected magnitude of the gradient does not change. It all depends on the step size.

  • @porimol108Tv
    @porimol108Tv 3 роки тому

    I subscribed to your channel.

  • @RichardMuenzer
    @RichardMuenzer Рік тому

    Question how does Newton play here?????

  • @aakarshrai5833
    @aakarshrai5833 5 місяців тому

    Bro could you please label you equations. It'll be helpful

  • @EdeYOlorDSZs
    @EdeYOlorDSZs Рік тому

    W explination

  • @theoutlet9300
    @theoutlet9300 4 роки тому

    dude where did you study this. understanding the maths make the coding so much fun

  • @yahavx
    @yahavx Рік тому

    Not enough formal for me.. intuition is nice but need to be a little concrete about what is actually done

    • @CodeEmporium
      @CodeEmporium  Рік тому +1

      Fair. I did what I cold in a short video like this. Thanks for watching!

    • @yahavx
      @yahavx Рік тому

      @@CodeEmporium Thank you!

  • @arsalan2780
    @arsalan2780 4 роки тому

    wonderfulllllllllllllll ...........!!!!

  • @Eysh2009
    @Eysh2009 5 місяців тому

    Valeu!

    • @CodeEmporium
      @CodeEmporium  5 місяців тому

      Thanks so much for the donation! Glad you liked this content!

  • @annarauscher8536
    @annarauscher8536 2 роки тому

    I think I watched that intro like 7 times haha

  • @EngRiadAlmadani
    @EngRiadAlmadani 4 роки тому

    Good jop

  • @stepantoman4694
    @stepantoman4694 2 роки тому

    hahahaha amazing

  • @MikeSieko17
    @MikeSieko17 8 місяців тому +1

    one critique, your notation is really weird and non intuative for beginners

  • @tostupidforname
    @tostupidforname 4 роки тому

    Imo you should have gone into more detail of the math of the optimizers. I did not understand how the terms relate to the behaviour the optimizers are supposed to have.

  • @artinbogdanov7229
    @artinbogdanov7229 4 роки тому +1

    Great explanation. Thank you!