AI/ML+Physics: Preview of Upcoming Modules and Bootcamps [Physics Informed Machine Learning]

This video provides a brief preview of the upcoming modules and bootcamps in this series on Physics Informed Machine Learning. Topics include: (1) Parsimonious modeling and SINDy; (2) Physics informed neural networks (PINNs); (3) Operator methods, like DeepONets and Fourier Neural Operators; (4) Symmetries in physics and machine learning; (5) Digital Twin technology; and (6) Case studies in engineering.
This video was produced at the University of Washington, and we acknowledge funding support from the Boeing Company
%%% CHAPTERS %%%
00:00 Intro & Recap
01:06 Reviewing the 5 Stages
04:08 Reviewing Physics in the Stages
05:11 Why Physical Models: Cost & Data Scale
07:53 Why Physical Models: Generalized Models
10:01 Why Physcial Models: Discovering Physics
11:40 Holistic Impact of Embedding Physics // Struggling to find a good wording here
12:55 Case Study: Pendulum Data and SINDy
15:20 Case Study: Symbolic Regression and Evolutionary Optimization
16:45 Case Study: Lagrangian Neural Networks
18:34 Architectures and Symmetries
19:36 Applications in Engineering
21:21 The Digital Twin
22:15 Benchmark Problems
23:35 Outro

[@drakkhein]

Steve and team is the perfect example of knowledge transfer in academics. His papers are really an enjoyable read + this channel is informative (and practical). Thanks, Steve!

[@ahmedhassaine3647]

I really appreciate what you are offering sir, ❤

[@weihongs5267]

thank you steve for share your knowledge, you are talent teacher

[@gebrilyoussef6851]

Prof Steve
Thanks for all the efforts you put to make this Bootcamp appears on UA-cam.
I wish if you can say something about "Geometric Deep Learning" and "Ricci Flow"

[@lukibakuoru876]

Can't wait for the next content. I have really learnt a lot

[@marco_burderi]

Fantastic Stuff! Can't wait.

[@mohammaddmour9127]

Waiting my dear teacher ❤❤❤❤

[@ESYoon-cf5eg]

Thanks for the great series of lectures, which I've been really enjoying. Around 12:00 and 13:00, it seems reference information for Raissi et al. and Lu et al. appears to be incorrect.

[@farquleetahmadkhan2710]

Anxiously waiting for it. Great #respect

[@antonispolykratis3283]

Looking forward...!!

[@bithigh8301]

When I grow up I want to teach like Steve!

[@integsoft]

Thank you Professor! Any time frame for the Digital Twin module?

[@LordMichaelRahl]

Fantastic as always. Will Bayesian optimization models and methods, e.g. for materials discovery, be covered as well?

[@idoben-yair429]

What about methods that integrate deep networks with classical methods, e.g., NN-as-preconditioner approach?
Thanks for great videos!

[@FreeeStorm]

Hello
Does anyone know if there is a discord or forum dedicated to physics-based machine learning?

[@alihan_ozturk]

please add kolmogorov arnold networks on series

[@alexanderskusnov5119]

Will there polyhedra and Geometric Algebra in the part 4 (symmetry)?

[@tikendraw]

I have been watching this videos as someone interested in data science . I really like these explainations, but in order to completely understand the workings we the viewers would like to see some code, some notebooks in action. DROP SOME TUTORIALS. thanks.

[@farquleetahmadkhan2710]

Please please actual problems more on each topic and thanks
#boeing

[@silverstreetman]

I can not wait to see the digital twins. When will you releaae that one doe digital twins?

[@jaikumar848]

Hello sir ! Could you please clear my basic doubt of control theory ? If system output in time domain defined by Y(t)= r(t) * H(t) then system output will be called unstable if there is any single value t exist for which Y(t) become infinite even though for other value of t system is constant? ?

[@elromulous]

There are different types of stability. E.g. asymptotically stable
en.wikipedia.org/wiki/Lyapunov_stability

[@arpitpatel5814]

You need to refer definition of stable system. Your answer is very much there.
Definitions are inescapable in science because they provide the foundational language needed to describe and understand phenomena accurately.

[@murithiedwin2182]

When current AI was post infancy to crawling, you were among the few people on the globe who made made us understand the concepts of latent Spaces and sparsity in AI and ML models as you taught us practical physics and engineering with maths.
Right now, you should be in tandem with the current state of advancements in AI. You should be previewing to us your models that do what you do with Physics, mathematics and ML, models that design dynamical systems all the way, from ground up to implementation all by themselves, while explaining and teaching just like you do...