An Introduction to Graph Neural Networks: Models and Applications

Поделиться
HTML-код
  • Опубликовано: 6 фев 2025

Комментарии • 127

  • @leixun
    @leixun 4 года назад +126

    *My takeaways:*
    1. Background 0:48
    2. Graph neural networks (GNN) and neural message passing 6:35
    - Gated GNN 26:35
    - Graph convolutional networks 29:27
    3. Expressing GGNNs as matrix operations 33:36
    4. GNN application examples 41:25
    5. Other models as special cases of GNNs 47:53
    6. ML in practice 49:28

    • @shawnz9833
      @shawnz9833 3 года назад +1

      can you help with that: what is a MLP in "you multiply it with a single layer MLP" @23:29?

    • @leixun
      @leixun 3 года назад +2

      @@shawnz9833 Multilayer perceptron

    • @shawnz9833
      @shawnz9833 3 года назад +1

      @@leixun Cool, thank you mate!

    • @leixun
      @leixun 3 года назад +1

      @@shawnz9833 You are welcome mate, I am doing some deep learning research as well. You are welcome to check out our research talks on my RUclips channel.

  • @rembautimes8808
    @rembautimes8808 3 года назад +4

    "Spherical Cow" - funniest analogy yet for a Neural Net layers. Great talk

  • @syllogismo
    @syllogismo 4 года назад +18

    Don't know why people are criticizing this video and the audience. Great introduction to graph neural networks!

  • @susmitislam1910
    @susmitislam1910 3 года назад +13

    So GNNs are basically something like calculating word embeddings in NLP. We have a dataset describing the relationships between pairs of words (nodes), and we want a vector representation that reflects how often they co-occur (weight of the edge between the nodes), i.e., how much relatedness the two words have. Once we have such vectors, we can build a vanilla, recurrent, or convolutional neural net to find out a mapping between the vectors and the output we desire.

  • @iltseng
    @iltseng 4 года назад +54

    At time 34:11, the (dot product of matrix A and matrix N) should be [ b + c ; c ; 0 ]

    • @MingshanJia
      @MingshanJia 4 года назад +3

      Actually, here what we use is the incoming edges (see 14:55), but that is true the slide is confusing about that.

    • @michaelkovaliov8877
      @michaelkovaliov8877 4 года назад +3

      It seems that the mistake is in the graph adjacency matrix, because the result vector is true given the drawing of the graph.

    • @PenguinMaths
      @PenguinMaths 4 года назад +19

      That is a mistake in the slide, A should be transposed to describe the incoming edges instead of the outgoing ones.

    • @khwajawisal1220
      @khwajawisal1220 4 года назад +1

      its using einstein notation, not the normal one that we use.

    • @kristofneys2349
      @kristofneys2349 4 года назад +3

      There are many mistakes or confusing comments in this presentation, no wonder the audience keeps asking questions. Not a good talk at all....

  • @alphaO27
    @alphaO27 3 года назад +2

    In 16:51, I think he meant for each node connecting to n (instead of n_j), because from the expression, we take all nodes n_j connected to n to be able to calculate the new state of node n h_t^n.

  • @16M-w4y
    @16M-w4y 9 дней назад

    Actually, I read the original research paper on this network, which established a new field in artificial intelligence. I found slight differences between the speaker's words and the research paper, then it became clear to me that he was talking in general about this neural network. In general, the audience's performance was unsatisfactory, and there were many interruptions that made the person confused and disorganized. The last thing I would like to say is that this field was found in the research paper of 2008, which was published in 2009. The researchers said that it is a network resulting from the features of RNN and its strengths and the idea of ​​working with Markov models, then it was wrapped in the concepts of graph theory.

  • @mehmetf.demirel8647
    @mehmetf.demirel8647 4 года назад +136

    great talk! the audience questions were helpful, but i felt like they were a bit too many in that they kinda negatively affected the flow of the talk.

  • @runggp
    @runggp 3 года назад +4

    awesome talk! The MSR audience asked quite a few questions, which are actually helpful , eg, what are they, how they work/update, why they are created and designed this way, etc

    • @shawnz9833
      @shawnz9833 3 года назад

      can you help with that: what is a MLP in "you multiply it with a single layer MLP" @23:29?

  • @ashutoshukey3803
    @ashutoshukey3803 3 года назад +27

    The explanation ability and use of high-level diagrams by the presenter were phenomenal. Questions from the audience definitely messed up the flow of the presentation quite a bit though.

  • @arjitarora8418
    @arjitarora8418 11 месяцев назад

    Great Introduction!

  • @rarelycomments
    @rarelycomments 17 дней назад

    This is why you have the Q&A at the end of the presentation.

  • @jonfe
    @jonfe Год назад +1

    I dont understand why GRU is used, the input in GRU is a (Node x Caracteristics) Matrix, where is the temporal dimension?

  • @heejuneAhn
    @heejuneAhn 4 года назад +3

    He explains using time progress, which make some cofusion to the audience and me.

  • @Peaceluvr18
    @Peaceluvr18 4 года назад +21

    Wow, what an excellent presentation, from someone with an ML background. Explains the basics a bit but also covers deep concepts. Super clear graphics! Seriously whoever made the graphics for this can I hire you to do my slide graphics? And thought it was very cool that the lecture attendees were bold enough to ask so many questions! Wish people asked more questions during my lectures+talks.

  • @Exhora
    @Exhora Год назад

    29:35 about CGNs, he said you multiply the sum of the messages with your own state. But in the equation, it is a sum. I didn't get which one is correct.

  • @soniagupta2851
    @soniagupta2851 2 года назад

    Excellent explanation

  • @MobileComputing
    @MobileComputing 4 года назад +87

    While the audience questions were mildly irritating (to put it, mildly), bombarding the speaker during his intro with questions that could reasonably be expected to be answered eventually in an 1-hour talk, why would the speaker give a talk on one of the most advanced neural network architecture to an audience without any machine learning background?

    • @ziranshuzhang6831
      @ziranshuzhang6831 4 года назад +15

      You are right, I am expecting to quickly adopt the GNN concept but the audience keeps asking irritating questions that I have to constantly hit the right button.

    • @ohmyumbrella
      @ohmyumbrella 4 года назад +1

      I agree. I mean I do see the point of giving this lecture to an audience without previous exposure to ML , if it is for the purpose of attracting them to the subject but in that case there should have been another video of the same lecture without so much interruption. It would take extra time and effort but for people who are trying to effectively learn GNN and have some knowledge of basic ML, these questions are very annoying and hinders the learning experience.

    • @robbat1209
      @robbat1209 6 месяцев назад

      The questions of the audience are absolutely valid, if they bother you there are plenty of other videos without an audience that you could watch

    • @MobileComputing
      @MobileComputing 6 месяцев назад +1

      @@robbat1209 This comment is absolutely valid. If it bothers you, there are plenty of other comments that you could read.
      This talk was from 4 years ago. This was one of the only sources about GNN, and before GenAI video summaries would allow audience like myself to comfortably skip ahead without fear of missing important information.

    • @robbat1209
      @robbat1209 6 месяцев назад

      @@MobileComputing 🤣🤣 true, my bad. The questions weren’t too horrible tho

  • @pielang5524
    @pielang5524 4 года назад +3

    Let the speaker talk!

  • @vritansh14
    @vritansh14 7 месяцев назад

    Amazin Talk !

  • @FsimulatorX
    @FsimulatorX 3 года назад +2

    OMG! For a second I thought he looked like the CEO of Google and was wondering to myself: why would the CEO of Google do a presentation about Neural Networks AT MICROSOFT!!

  • @codewithyouml8994
    @codewithyouml8994 3 года назад

    The miss, at 40:00 was right .... as i was alsoooo really confused, like all the matrix operations were seemed to be invalid if not swapped ... lol what kind of inverted conventions are these ....

  • @mansurZ01
    @mansurZ01 4 года назад +3

    35:40 I think the dimensionality of M should be (num_nodes x D), unless D==M.
    EDIT: from what follows, it should be M = HE, and D can be different from M.

  • @instantpotenjoyer
    @instantpotenjoyer 4 года назад +14

    DiscussIon on the actual topic starts ~ 6:40

  • @peter-holzer-dev
    @peter-holzer-dev Год назад +1

    I am pretty sure this is a great talk but unfortunately all the questions in between disturbs the flow a lot (also because most of them are hard to understand acoustically).

  • @lidiias5976
    @lidiias5976 7 месяцев назад

    36:53 what is M in the shape (num_nodes by M)?

  • @eljangoolak
    @eljangoolak 3 года назад +8

    Very good presentation, but it is very difficult to follow with all the interrupting questions

  • @sunaxes
    @sunaxes 3 года назад

    So GNN is just message passing on a graph or did I miss something? This has been around since way back, isnt it??

  • @samuisuman
    @samuisuman 4 года назад +4

    There is no intuitive explaination, but quite informative

  • @aladdin1776
    @aladdin1776 7 месяцев назад

    The talk is very interesting; however, the interruptions from the audience are quite disturbing.

  • @leventesipos1456
    @leventesipos1456 5 месяцев назад

    The notation is incomplete or incorrect in so many places on the presentation, that it was hard to follow.

  • @mohammedamraoui4147
    @mohammedamraoui4147 11 месяцев назад

    and for Edge classification?

  • @klaimouad740
    @klaimouad740 3 года назад

    can we propagate messages for example depending on the edges features for example if the distance from node n to m is greater than their distance to p then we propagate the message first to p then we perform the propagation to the other node m

  • @pharofx5884
    @pharofx5884 4 года назад +91

    dat cough frequency suspiciously high. Distance thyself socially sir.

    • @maloxi1472
      @maloxi1472 4 года назад +5

      Don't worry, it's from November 2019

    • @jmplaza4947
      @jmplaza4947 4 года назад +4

      @@maloxi1472 COVID was already spreading then, right? I hope he's ok... wherever he is now...

    • @danielliu9616
      @danielliu9616 4 года назад +5

      He is maybe THE patient zero

    • @yanzexu6912
      @yanzexu6912 4 года назад

      bless him

    • @stackexchange7353
      @stackexchange7353 4 года назад

      @@danielliu9616oh shii

  • @NewtonInDaHouseYo
    @NewtonInDaHouseYo 3 года назад

    Excellent introduction, thanks a lot!

  • @arisioz
    @arisioz 2 года назад +4

    Are those actual MS employers in the crowd? They are worse than 1st year CS students

  • @DED_Search
    @DED_Search 3 года назад +5

    is there anyway to get access to the slides? Great talk! Thanks.

  • @hfkssadfrew
    @hfkssadfrew 3 года назад +2

    34:46 . it is NOT A* N, it is N' * A ....

  • @sherazbaloch1642
    @sherazbaloch1642 2 года назад

    Need more tutorial on GNN

  • @sm_xiii
    @sm_xiii 4 года назад +5

    Where can we get the slide deck please?

  • @halilibrahimakgun7569
    @halilibrahimakgun7569 Год назад

    Can you share the slides please. I like them.

  • @losiu998
    @losiu998 3 года назад +3

    Great presentation. But I have to point out something. I have no idea why you would use einstein notation instead of simple matrix multiplication? It raises unnecessary confusion and it's not related
    to GNNs.

  • @r.dselectronics3349
    @r.dselectronics3349 4 года назад

    i am a researcher..the video contain the beautiful concept ...i like very much....:)..specially binary classification part ...i am so excited about this concepts....

  • @goblue1011
    @goblue1011 2 года назад +3

    Honestly some of the audience who raised questions have quite big ego and have no idea what they are talking about.

  • @hamishhall5423
    @hamishhall5423 4 года назад

    What is the dimension M, for the msg_to_be_sent and received_messages etc. I get that D is the dimension of the node representation, N the num_nodes etc

  • @pielang5524
    @pielang5524 4 года назад

    at time 35:51, the adjacency matrix A should also depend on edge type k imo.

    • @pielang5524
      @pielang5524 4 года назад

      OK.. The presenter confirmed this shortly after...

  • @yb801
    @yb801 3 года назад

    Is this a 2016 talk?

  • @sebamurgui
    @sebamurgui 4 года назад +49

    OH MY GOD that audience

    • @miketurner3461
      @miketurner3461 4 года назад +4

      I had to stop watching because of them

    • @kellybrower301
      @kellybrower301 4 года назад

      I wanted to believe it wasn’t true.

  • @BapiKAR
    @BapiKAR 4 года назад +2

    could you please post the ppt here? Thanks

  • @giannismanousaridis4010
    @giannismanousaridis4010 4 года назад +2

    I found the slide for everyone who asked here: miltos.allamanis.com/files/slides/2020gnn.pdf
    (Idk If I'm not supposed or allowed to post the link here, if not sorry for that, I'll delete my comment. Just let me know).

    • @warpdrive9229
      @warpdrive9229 7 месяцев назад

      You can let it remain here :)

  • @2000sunnybunny
    @2000sunnybunny 4 года назад

    Great session !

  • @lifelemons8176
    @lifelemons8176 3 года назад

    horror crowd. this is something I see in every microsoft talk

  • @minghan111
    @minghan111 4 года назад +23

    too many questions, just wait for the speaker please.

    • @pharofx5884
      @pharofx5884 4 года назад +6

      In formulating their questions they re-explained what is going on an order of magnitude better than the speaker. thats kinda sad

    • @MobileComputing
      @MobileComputing 4 года назад +1

      @@pharofx5884 I just watched the version from 2 years ago. Only 18 minutes long, but almost identical in content, yet that was much clearer. Really sad to see.

  • @arashjavanmard5911
    @arashjavanmard5911 4 года назад +4

    It would be great if you could also publish the slides!

    • @mayankgolhar8761
      @mayankgolhar8761 4 года назад +5

      Slides from the presenter's website: miltos.allamanis.com/files/slides/2020gnn.pdf

    • @arashjavanmard5911
      @arashjavanmard5911 4 года назад +1

      @@mayankgolhar8761 thanks

  • @CodingwithIndy
    @CodingwithIndy 4 года назад

    Is that SPJ I hear in the audience at 9:18

  • @RARa12812
    @RARa12812 3 года назад

    How tp turn off questions

  • @WeeeAffandi
    @WeeeAffandi Год назад

    You can tell this lecture was recorded during the prime Covid by hearing the constant coughing from audience (and the speaker)

  • @BruceChen27
    @BruceChen27 4 года назад +5

    Seems several people were not healthy

  • @aichemozenerator8446
    @aichemozenerator8446 2 года назад

    good

  •  4 года назад +1

    Where is the inputs and outputs?

    • @miketurner3461
      @miketurner3461 4 года назад

      Clearly you would've been one of the people asking foolish questions they could answer using Google

  • @losiu998
    @losiu998 3 года назад

    34:22 - is it vector-matrix multiplication? if so, the result is wrong i guess
    @edit: Matrix A should have ones under diagonal, not above - then result is as presented

  • @ZephyrineFreiberg
    @ZephyrineFreiberg 3 года назад +1

    It should be (A^T)*N

  • @carloderamo
    @carloderamo 2 года назад +1

    Some good person should take this video and remove all the awful questions from the audience

  • @blanamaxima
    @blanamaxima 3 года назад +1

    Is this related to bread baking ?

  • @sirisaksirisak6981
    @sirisaksirisak6981 3 года назад

    At least it save time in doing strategy.

  • @kutilkol
    @kutilkol 3 года назад +2

    Rip ears. Wtf with the caughing. Use at least some compressor for the vocal audio omg.

  • @patrickadjei9676
    @patrickadjei9676 4 года назад +3

    People are not happy for the many questions. However, I'm kinda sad that he doesn't re-state the questions before answering :( like why?

  • @ngideo
    @ngideo 4 года назад +1

    I hear Simon Peyton Jones in the audience

    • @rherrmann
      @rherrmann 3 года назад

      Easily recognizable indeed!

  • @khwajawisal1220
    @khwajawisal1220 4 года назад +8

    thats why software engineers should not teach, you assume everything is a design/modeling detail when in reality they are part of the mathematics behind them. and i seriously miss those old days where professors used to teach with a chalk and board.

  • @guest1754
    @guest1754 Год назад

    Wondering how many people had covid in that recording...

  • @rocking4joy
    @rocking4joy 3 года назад +1

    Don't understand the praise in the comment section, I actually found it kind of sloppy with typo-s but the audience and the questions are really great.

  • @dalahmah
    @dalahmah 7 месяцев назад

    I found the lecture’s atmosphere dull and depressing. It seems that the lecturer was forced to give the lecture!

  • @JK-sy4ym
    @JK-sy4ym 8 месяцев назад

    Unfortunately many good researchers can’t present their work well.

  • @yanyipu4029
    @yanyipu4029 3 года назад

    Great Video but annoying audience

  • @jebinjames6317
    @jebinjames6317 3 года назад

    People need to remember they’re watching a free video on RUclips...it’s not your advanced ML private tutoring session...

  • @kushalneo
    @kushalneo 3 года назад

    👍

  • @KeshavDial
    @KeshavDial 3 года назад

    The audience ruined this presentation. I have never felt worse for a presenter.

  • @thevanisher4609
    @thevanisher4609 Год назад

    Horrible audience, great talk!

  • @aglaiawong8058
    @aglaiawong8058 3 года назад

    the interruptions are so annoying...

  • @ММ_4321
    @ММ_4321 3 года назад

    Kind of confusing for me. And the audience very annoying

  • @yiweijiang
    @yiweijiang 4 года назад +1

    So that's machine learning! Haha, lol

  • @barriesosinsky9566
    @barriesosinsky9566 3 года назад

    An aromatic ring is not a "single bone" next to a "double bone." The bonds are a resonance form in a single state. Treating them with graph theory is not supported by current models.

  • @iSpades0
    @iSpades0 3 года назад

    that audience was pretty annoying tbh

  • @williamashbee
    @williamashbee 4 года назад

    cough totally ruined the presentation.

  • @eaglesofmai
    @eaglesofmai 3 года назад

    are GNN's patented? does anyone know if using a paritcular ANN construct can be subject to litigation?

  • @MrArmas555
    @MrArmas555 4 года назад

    ++

  • @miketurner3461
    @miketurner3461 4 года назад +2

    The audience needs to take a freaking ML 101 class before asking stupid questions

  • @robodoc1446
    @robodoc1446 3 года назад

    Appalling talk! It shows why coders are terrible in public speaking or often fail to explain things in a transparent manner. Before explaining how message passing is done in an end-to-end learning architecture, he jumped to talk about Gated GNN leaving an impression that GRU may be an important part of GNN. This is one of the reasons why he got so many questions and confusion surrounding his lecture.....what is h_t? "well, this is not something that changes"... seriously Microsoft!

  • @Tyomas1
    @Tyomas1 4 года назад +2

    awful introduction

  • @David_Lloyd-Jones
    @David_Lloyd-Jones 2 года назад

    Word salad. A hopeless mess of talking at and around a topic without actually touching it.
    Take it down. Tell the guy to try again.

  • @saulrojas2679
    @saulrojas2679 2 года назад

    Slides can be found at: miltos.allamanis.com/files/slides/2020gnn.pdf