Equivariant Neural Networks | Part 3/3 - Transformers and GNNs

▬▬ Papers / Resources ▬▬▬
SchNet: arxiv.org/abs/1706.08566
SE(3) Transformer: arxiv.org/abs/2006.10503
Tensor Field Network: arxiv.org/abs/1802.08219
Spherical Harmonics UA-cam Video: • What are Spherical Har...
Spherical Harmonics Formula: • The spherical harmonics
Tensor Field Network Jupyter Notebook: github.com/UPEIChemistry/tens...
SE(3) Repo: github.com/FabianFuchsML/se3-...
NVIDIA Updated Version: developer.nvidia.com/blog/acc...
▬▬ Used Music ▬▬▬▬▬▬▬▬▬▬▬
Music from #Uppbeat (free for Creators!):
uppbeat.io/t/yokonap/birds
License code: WXVHOOZRRWDUCKIU
▬▬ Used Icons ▬▬▬▬▬▬▬▬▬▬
All Icons are from flaticon: www.flaticon.com/authors/freepik
▬▬ Timestamps ▬▬▬▬▬▬▬▬▬▬▬
00:00 Introduction
00:43 Points, Graphs and Sets
01:11 Inductive Biases & Equivariance
03:15 3D is not commutative
04:38 SchNet
05:48 Tensor Field Networks
07:17 Math Terminology
12:47 Hands on TFNs
13:08 SE(3) Transformer
15:24 Hands on SE(3) Transf
▬▬ Support me if you like 🌟
►Link to this channel: bit.ly/3zEqL1W
►Support me on Patreon: bit.ly/2Wed242
►Buy me a coffee on Ko-Fi: bit.ly/3kJYEdl
►E-Mail: [e-mail]
▬▬ My equipment 💻
- Microphone: amzn.to/3DVqB8H
- Microphone mount: amzn.to/3BWUcOJ
- Monitors: amzn.to/3G2Jjgr
- Monitor mount: amzn.to/3AWGIAY
- Height-adjustable table: amzn.to/3aUysXC
- Ergonomic chair: amzn.to/3phQg7r
- PC case: amzn.to/3jdlI2Y
- GPU: amzn.to/3AWyzwy
- Keyboard: amzn.to/2XskWHP
- Bluelight filter glasses: amzn.to/3pj0fK2

[@chenweilong2505]

Incredible video. Awesome! Thanks!

[@IgorItkin]

Thanks man!

[@v_pryadchenko]

Cool!

[@IMdigitalworks]

Hey, thank you for the video. Do you know some resources about SE(3) invariant gnns for points clouds? That is, I would like a gnn to encode local geometries of point neighborhoods that be would invariant under SE(3) (not equivariant).

[@faiazahsan6774]

Your GNN video serise is really helpful for a beginer like me. Thank you very much.
p.s: Any chance you can make a project video on Supply Chain Risk Detection using GNN?

[@tilkesh]

Thx

[@v_pryadchenko]

Where could I find links to parts 1 and 2?

[@julienblanchon6082]

Part1: ua-cam.com/video/2bP_KuBrXSc/v-deo.html
Part2: ua-cam.com/video/r0xyxe31QgU/v-deo.html

[@prajwol_poudel]

Do you have a background in mathematics?
coming from a non-mathematics, non-computer engineering/ CS background, the mathematics I've learned so far feels like cave-man mathematics when seeing all the fancy mathematics in the video, can't even imagine trying to read any of the paper on my own.
How can someone without a good background in mathematics navigate/dive into such topics which demands high mathematical rigour?

[@DeepFindr]

Hi! Not directly mathematics, but computer science. Yeah I agree I doesn't look so trivial but I just read different articles about it until I mostly understood it. But I'm far from being an expert and there are still many nuances I didn't fully understand.
Took me ages however ;-) This area is certainly one of the math heavier ones. I think given enough time it is possible to understand such papers, but the question is if it's really worth to invest it.
The next videos will be more comprehensible again :)

[@zeyutang2084]

@@DeepFindr Hi thanks for the amazing series! I am from an engineering background, just wondering if you could recommend any books or lectures for group theory since I would like to develop a more rigorous mathematical understanding of the topic. Many thanks