🧠 Scientific Machine Learning, FEM + ML, PINNs – Ehsan Haghighat | Podcast #79
YouTube Viewers YouTube Viewers
32.5K subscribers
5,282 views
0

 Published On Aug 3, 2022

🧠 Science Academy: https://courses.jousefmurad.com/
🌎 Website: http://jousefmurad.com

Dr. Ehsan Haghighat is a Postdoctoral Fellow at UBC studying stochastic modeling and uncertainty quantification of engineering systems. Previously, he was a Postdoctoral Associate at MIT where he studied the assessment of induced seismicity due to CO2 sequestration and oil and gas injection and production, Stochastic Modeling, and Machine Learning.

He received his Ph.D. from McMaster University specializing in Computational Geomechanics. His research interests include computational methods for the mechanics of solids and porous media, stochastic modeling and uncertainty quantification, and machine learning of engineering systems.

ONLINE PRESENCE
================
🌍 My website - http://jousefmurad.com/
💌 My weekly science newsletter - https://jousef.substack.com/
📸 Instagram -   / jousefmrd  
🐦 Twitter -   / jousefm2  

SUPPORT MY WORK
=================
👕 Science Merch: https://engineered-mind.creator-sprin...
👉 Subscribe to my Science Blog: https://www.jousefmurad.com/#/portal/...
👉 Become my Patreon:   / theengiineer  

🧠 Subscribe for more free videos: https://bit.ly/2RLmMxq

RECOMMENDATIONS
==================
🎥 1 Month Free of Skillshare Premium: https://bit.ly/3E3bXwC
📕 Books on AI, Data Science & Programming: https://bit.ly/2I1lUD1 - use "podengineered20" to get a 40% discount!

CONTACT:
-----------------
If you need help or have any questions or want to collaborate feel free to reach out to me via email: [email protected]

#pinns
#simulation
#machinelearning

Disclaimer: Some of these links are affiliate links which make me earn a small commission when you make a purchase at no additional cost to you.

TIME STAMPS
-----------------------
0:00 : Intro
6:20 : How to combine FEM + ML
13:04 : SciAnn
21:19 : Output of the NN
22:05 : Can someone adapt the output parameters?
23:14 : PINN vs. Classic Approach - TIme Saving
27:16 : Why PINNs?
31:09 : SciAnn - A Black Box?
33:06 : XAI for PINNs
34:32 : SciAnn in the Future
38:58 : Ehsan's other projects
48:37 : FEM & CFD - Transfer Learning Using Only Weights?
51:54 : SciAnn - No Big Workstation Needed :)
53:14 : Resources Ehsan Uses to Stay Up-To-Date
55:01 : Closing Remarks


Podcast Recorded: August, 11th 2021 - Subscriber Release Count: 18,466.

show more

Share/Embed