Self-/Unsupervised GNN Training

▬▬ Papers/Sources ▬▬▬▬▬▬▬
- Molecular Pre-Training Evaluation: arxiv.org/pdf/2207.06010.pdf
- Latent Space Image: arxiv.org/pdf/2206.08005.pdf
- Survey Xie et al.: arxiv.org/pdf/2102.10757.pdf
- Survey Liu et al.: arxiv.org/pdf/2103.00111.pdf
- Graph Autoencoder, Kipf/Welling: arxiv.org/pdf/1611.07308.pdf
- GraphCL: arxiv.org/pdf/2010.13902.pdf
- Deep Graph Infomax: arxiv.org/pdf/1809.10341.pdf
- InfoGraph: openreview.net/pdf?id=r1lfF2NYvH
▬▬ Used Icons ▬▬▬▬▬▬▬▬▬▬
All Icons are from flaticon: www.flaticon.com/authors/freepik
▬▬ Used Music ▬▬▬▬▬▬▬▬▬▬▬
Music from Uppbeat (free for Creators!):
uppbeat.io/t/mountaineer/autu...
License code: EHK2BNGUHRZX1CXK
▬▬ Timestamps ▬▬▬▬▬▬▬▬▬▬▬
00:00 Introduction
00:20 Applications
01:48 Unsupervised Learning on Graphs
02:28 Self-Supervised Learning
03:19 Overview SSL on Graphs
04:44 Graph/Node-level
05:16 Reconstruction Approaches
07:10 Predictive / Task Generation Approaches
08:12 Contrastive Approaches
10:51 Comments on Similarity
11:30 Final Remarks
▬▬ Support me if you like 🌟
►Website: deepfindr.github.io/
►Support me on Patreon: bit.ly/2Wed242
►Buy me a coffee on Ko-Fi: bit.ly/3kJYEdl
►Coursera: imp.i384100.net/b31QyP
►Link to this channel: bit.ly/3zEqL1W
►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

[@jihochoi_cs]

Amazing video! Thanks for the great work.

[@radhen171992]

This video is exactly what I was looking for! Great summary.

[@solo_driven]

Great channel! Keep up the good work!

[@NeotenicApe]

You are incredible

Great as always :D

[@MaryamKamali-ln4lz]

Thanks for really great videos! Could you please create a video on temporal variational graph autoencoders?

[@sakib.9419]

bro your channel is insane!

[@DeepFindr]

Haha thanks! Glad you like it

[@sakib.9419]

@@DeepFindr I do a lot of machine learning for my work and I’ve recently thought about modelling train networks as a graph to apply algorithms. d’you have a discord or something, would love to chat a bit?

[@DeepFindr]

Hi! I'd love to but unfortunately I don't find time to maintain a discord or anything like that. Maybe in the future :)

[@RishabhSingh-hc4nz]

Ur work is appreciable sir pls upload next video and also make the movie recommender system code explanation

[@trez6465]

Amazing video. What tool do you use to make the slides and the beautiful graphics?

[@DeepFindr]

Thanks :) DaVinci resolve and PowerPoint :P

[@OGIMxGaMeR]

Do you know any libraries implementing self supervised learning on graphs?

[@mrinmoybanik5598]

Hmm can make something on solving NP hard combinatorial graph problems using GNNs?

[@comp-it1913]

I am looking forward to present a lecture based in dynamic graphs variational autoencoder where temporal dependencies occured in dataset such as autonomous driving for point clouds change at each time instant,also to get latent space Z. Please read a paper joint dynamic variational graph autoencoder and make a video on its implementation step by step. I will be very thankful

[@user-lb9fv3kd8v]

You can search
VGRNN(Variational Graph Recurrent Neural Network) which is handling dynamic graph by snapshot. It means that sequences graph (=dynamic graph) can see multi-static graph. This method called discrete method.

[@dzxxbj]

Can show some demos on google colab, are there any more applications ?

[@DeepFindr]

Will do a contrastive tutorial with code soon :) what kind of applications do you mean?

[@ghadaalmousa9248]

Hello,
Do you think Generative Adversarial Networks can improve the performance of GCN . if that so how it work?/

[@washedtoohot]

Yes

[@washedtoohot]

No

[@user-vx4tc6pr9l]

@@washedtoohot so yes or no ?