Відео

As a Chemical Engineer that studied CFD in grad school turned Data Scientist, I absolutely love this and the fact that there is active research in the intersection of physics and AI.

what id really mass energy and momentum created then how will gauss divergence theorem work?

This is awesome! I am so stocked about how they have found a 15D linear system with integer sparse off-diagonal matrix approximating Lorenz attractor with such an accuracy. It is a pure joy contemplating this mathematical elegance

It's not just a log function that has levels. The numbers themselves have levels. exp(i t) = exp(i (t + 2 pi k)) for all integer k It seems to me that any complex valued function also has levels. Why is the Log function singled out as special? Is it because it has a singularity?

epic lighting. subscribed.

Your words of choice are easy to understand and make the lecture sounds fun! Thankyou! Will sit for my final this 29june pray for me🙏

9:21 so you change the lower bound to 0 because the lowest value we can obtain from solving the integral is now 0, rather than negative infinity, since you've defined it to be that way using H(t)? Is that about right?

You outdid yourself as usual

Amazing, Than you

Amazing, Thank you

2024, still the best.

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

STEVE I ABSOLUTLY LOVE HOW SHARP AND CLEAN AND PACKED FULL OF INFO YET STILL DIGESTIBLE EVERYTHING IS!! You are my favorite teacher. I am a self learner and you are GOD MODE for that. PS: please do a video on why all power series are Taylor series (without borel heavy machinery if possible or by using borel but breaking everything down)!

Why is EVERY power series a Taylor series (without having to use heavy analysis stuff I don’t understand)!?

You know that it is going to be a great lecture series when Eigensteve is teaching you about eigenvalues and eigenvectors

tbh understanding his videos is very difficult, IMO he explains badly. Like 14:14 is the first more complicated part and I don't really get what it is about. I wouldn't understand ResNet from his explanation either if I had no prior knowledge about it. He just assumes that I am some expert in math and DLs

suposse we have a discrete time signal x[n]=exp(jwn) and it is periodic with N. then exp(jw(n+N))=exp(jwn) thus exp(jwN)=1. because 1=exp(2(pi)k) where k is an integer, equation w=2(pi)k/N must hold. if N is chosen to be pi ,which is not an integer, x[n] is not periodic. consequently, has infinitely many unique values. In addition, for x[n] to be periodic, w must be some multiple of pi (true when k and N are integers).

I'm so much very grateful for these videos you make. Keep on the good work.

This is so technically correct, and simultaneously so obtuse, that my intuition fuse has melted. Please consider redoing this as 3D pseudo visualizations of data subsets.

tim cook?

How can sigma be invertible when it's a rectangular matrix?

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.

this video is on a different level of teaching, thanks

Nice vid and looking forward to follow all the content. By the way, how physics informed ML differs from the field of system identification?

THANK U SIR . THAT WAS EXACTLY WHAT I AM LOOKING FOR.

Thank you Steve. Great explanation! Greets from Argentina!

18:50 also setting Ki >> Kp for the PI control shows the oscillation around the reference due to the intergral term very nicely

Can someone please edit the squeaks out. I was really hoping to watch the videos. But the squeaks are jarring.

Thanks Prof. Brunton, I really like your videos. My question is what if we have the system's data but it is not the impulse experiment. It is a combination of step response and chirp response. What should we do? I don't have access to the system anymore. Any input is appreciated.

Amazing as usual

Professor please show me that when a unit mass as a wave propagate and transfer energy to the mass energy is kept constant. I can find particle velocity and shear strain for a shear wave and the displacement at a particular point for any time t but I don’t get the total energy of at the point does not main the same value. As shear strain is directly related to the particle velocity, is it that I have to consider either particle velocity or shear strain plus displacement related velocity in the perpendicular direction of displacement. Please help me.

You have an impressive ability to simplify complex subjects

Excellent explanation! For those interested in LLMs residual connections are also featured in the vanilla transformer block. The idea is similar to CNN ResNets, but instead of gradually adding pixel resolution each block adds semantic "resolution" to the original embedded text input.

IMHO you should speak a little faster and make less jests; for scientists watching this it wastes a time and attention.

Thank you a lot!

amazing as usual, I wish I had discovered this channel long time ago, control systems is my favorite field and this channel made me even love it more, thank you for clear explanation!

Does anyone know if transformers have superseded resnets for image classification?

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

Hi Steve - this is Vivek Karmarkar! Thanks for the video - great content as usual and keeps me motivated to create my own PINN content as well. Looking forward to the next video in the series and would love to talk PINN content creation with you! I have been thinking about presenting PINNs with ODEs as examples and its nice to contrast it with Neural ODEs - nomenclature aside, it looks like the power of the NN as universal approximators allows us to model either the flow field (Neural ODEs) or the physical field of interest (PINNs) for analysis which is pretty cool!

Many thanks for the extremely good videos. Really well explained and easy to understand. A video on "Augmented neural ODEs" would go well with "neural ODEs" 😊

ResNet is literally the best thing happened in Deep Learning.

9:27- intro another entity here really messed things up

RtoL writing behind a glass is revolutionizing online teaching, the 1st of which seems to be that on Nancy's channel on calculus. Super cerebrum-cerebellum-hand axis👍

Thank you very much for your videos. I am glad that besides the classical sources to promote science (such as books and papers), your lectures can also be found on youtube. In my opinion, Prof. Bruton is the best provider of youtube lectures and I don't want to miss any of the lectures.

thank u sire!

SHAMEEEEEE🎉 Bound, border, infinity, noninfinity, natural, where is the end?! calc machine how it works, integer, costs money costs profits cons in mathematics, NOMIA! ECO? algorithm accuracy, fuzzy logic, integer 0-I. ONE BOOK NO INDIVIDUAL HERE 🎉WHEN YOU SMOOTHING GRADIENT YOU LOSING

Thank you for making this. I watched your video on Neural ODEs before I watched this. It is much easier to understand the Neural ODE video now that I have watched this. I would love to watch a video about the ResNet classifier idea you discuss here. If you have already done that please add a link here. Thanks, and this is awesome!

Thank you SO MUCH prof. Steve!! I intend to study neural networks in my graduate courses thanks to your lectures!! :D

STEVE MAKE TWO. SMALLER HIGHER LIKE ARRAY ONE DIRECTION OR SYMMETRY LIKE MIRROR. FEEDBACK AND THIS 150.000ageSCIENCE.

Thank you for this content! Love your approach. Please never stop educating people. We all need teachers like you!:) ps Enjoying reading your book