Convolution

Outline

1. Convolutional Neural Networks - quick intro

2. Convolutional Neural Networks - more details

3. Graph Convolution Basics Applied to Molecules

Videos

Ideas for further coding projects 

• Is the solubility data used in video 4 the same as the one you used for multi-linear regression?

• Are there any outliers in video 4? What do they look like?

• How important are each of the descriptors used in video 4?

• Can you figure out how to change the code in video 4 from a classification code to a regression code?

Inspiration/Further Reading

https://victorzhou.com/blog/intro-to-cnns-part-1/

https://victorzhou.com/blog/intro-to-cnns-part-2/

https://victorzhou.com/blog/keras-cnn-tutorial/

https://towardsdatascience.com/how-to-do-deep-learning-on-graphs-with-graph-convolutional-networks-7d2250723780

https://towardsdatascience.com/how-to-do-deep-learning-on-graphs-with-graph-convolutional-networks-62acf5b143d0

https://iwatobipen.wordpress.com/2019/04/05/make-graph-convolution-model-with-geometric-deep-learning-extension-library-for-pytorch-rdkit-chemoinformatics-pytorch/

https://arxiv.org/pdf/1609.02907.pdf

https://tkipf.github.io/graph-convolutional-networks/

https://www.youtube.com/watch?v=LFSR27BaNIQ

https://www.youtube.com/watch?v=0_O8PdZBc5s (talk by Welling)