Neural ODEs (NODEs) [Physics Informed Machine Learning]

Рет қаралды 54,452

Күн бұрын

This video describes Neural ODEs, a powerful machine learning approach to learn ODEs from data.
This video was produced at the University of Washington, and we acknowledge funding support from the Boeing Company
%%% CHAPTERS %%%
00:00 Intro
02:09 Background: ResNet
05:05 From ResNet to ODE
07:59 ODE Essential Insight/ Why ODE outperforms ResNet
// 09:05 ODE Essential Insight Rephrase 1
// 09:54 ODE Essential Insight Rephrase 2
11:11 ODE Performance vs ResNet Performance
12:52 ODE extension: HNNs
14:03 ODE extension: LNNs
14:45 ODE algorithm overview/ ODEs and Adjoint Calculation
22:24 Outro

Пікірлер: 36

@smustavee Ай бұрын

I have been playing with NODEs for a few weeks now. The video is really helpful and intuitive. Probably it is the clearest explanation I have heard so far. Thank you, Professor.

@mohammadxahid5984 2 ай бұрын

Thanks Dr. Brunton for making a video on Neural ODE. Came across this paper as soon as it came out back in 2018. Still goes over my head particularly the introduction of the 2nd differential equation/ adjoint sensitivity method. Would really appreciate if you explain it in detail.

@astledsa2713 Ай бұрын

Love your content ! Went through the entire complex analysis videos, and now gonna go through this one as well !

@stefm.w.3640 14 күн бұрын

Great video, I learned a lot! Piqued my interest and inspired me to do a deep dive into all the topics mentioned

@as-qh1qq Ай бұрын

Amazing review. Engaging and sharp

@anthonymiller6234 Ай бұрын

Awesome video and very helpful. Thanks

@lucynowacki3327 10 күн бұрын

Cool summary and intro for liquid NNs.

@daniellu9499 Ай бұрын

very interesting course, love such great video...

@kepler_22b83 Ай бұрын

So basically rising awareness that there are better approximations to "residual" integration. Thanks for the reminder. From my course on numerical computation, using better integrators is actually better than making smaller time steps, rising the possible accuracy given some limited amount of bits for your floating point numbers.

@codybarton2090 Ай бұрын

I love it great video

@joshnicholson6194 Ай бұрын

Very cool!

@OnionKnight541 10 күн бұрын

this is great --- i think about this stuff all the time, but didn't know others did :/

@hyperplano Ай бұрын

So if I understand correctly, ODE networks fit a vector field as a function of x by optimizing the entire trajectory along that field simultaneously, whereas the residual network optimizes one step of the trajectory at a time?

@topamazinggadgetsoftrendin2916 Ай бұрын

Very interesting

@merrickcloete1350 Ай бұрын

@Eigensteve is the nth order runge kutta integrator not just what a UNet is, after its being properly trained. The structure appears the same and the coefficients would be learned.

@HD-qq3bn Ай бұрын

I study neural ode for quite a long time, and found it is good for initial value problem, however, for external input problem, it is really hard to train.

@ricardoceballosgarzon6100 Ай бұрын

Interesting...

@osianshelley3312 23 күн бұрын

Fantastic video! Do you have any references for the mathematics behind the continuous adjoint method?

@etiennetiennetienne Ай бұрын

I would vote for more details on the adjoint part. It is not very clear to me how to use AD for df/dx(t) now that x changes continuously (or do we select a clever integrator during training?) .

@SohamShaw-bx4fq Ай бұрын

Can you please teach latent neural ode in detail?

@-mwolf Ай бұрын

Awesome video. One question I'm asking myself is: Why isn't everybody using NODEs instead of resnets if they are so much better?

@smeetsv103 Ай бұрын

If you only have access to the x-data and numerically differentiate to obtain dxdt to train the Neural ODE. How does this noise propagate in the final solution? Does it acts as regularisation?

@The018fv Ай бұрын

Is there a model that can do integro-differential equations?

@Heliosnew Ай бұрын

Nice presentation Steve! I just gave a very similar presentation on Neural ODE-s just a week prior. Would like to see it one day to be used for audio compression. Keep up the content!

@marcelotoledo1820 19 күн бұрын

Why is it implicit that x(k+1)=x(k)+f(x) is Euler integration ? Can be any integrator depending on how you build f(x), Runge Kutta for example f is f(x) =h/6*(k1+2*k2+2*k3+k4).

@franpastor2067 20 күн бұрын

What about periodic functions? Is there a way to get nice approximations with neural networks?

@zlackoff Ай бұрын

Euler integration got dumped on so hard in this video

@sucim 15 күн бұрын

Very confusing presentation! First Neural ODEs are presented as a continuous version of ResNets, which would imply that the integration happens in "depth" which would make them similar to fully-connected or convolutional neural networks (non-sequence models). The afterwards it is suggested that the integration actually happens in "time" which makes neural ODEs much more similar to sequence models. Even ChatGPT et al. are confused and can't answer this distinction properly. Seems like it is a quite buzzword-driven field...

@digriz85 Ай бұрын

Nice video, but I really miss the connection point between the NNs and the math part. I have a PhD in physics and I've worked a lot with the math you're talking about. Also I've worked a few years as a data scientist and I kinda understand how it goes with the neural networks. But I really miss the point how you make these two work together. Sorry if I sound dumb here.

@anonym9323 Ай бұрын

Does some one have a example repository or libary so i can plaz with it

@devinbae9914 Ай бұрын

Maybe in the Neural ODE paper?

@edwardgongsky8540 Ай бұрын

Damn I'm still going through the ode and dynamical systems course, this new material seems interesting AF though

@user-oj9iz4vb4q 25 күн бұрын

This seems like you are changing your loss function not your network. Like there is some underlying field you are trying to approximate and you're not commenting on the structure of the network for that function. You are only concerning yourself with how you are evaluating that function (integrating) to compare to reality. I think it's more correct to call these ODE Loss Functions, Euler Loss Functions, or Lagrange Loss Functions for neural network evaluation.

@erikkhan Ай бұрын

Hi Professor , What are some prerequisites for this course?

@tramplerofarmies Ай бұрын

I suspect these are not the type of courses with defined prereqs, but def need calculus series, linear algebra series, and some computer science. To really understand it, classical mechanics and signals and systems (control theory, discrete and continuous).