Week 40 - October 21

Monday 10/4

Jupyter Lab is now packaged as a desktop app. Gave a try 2 minutes but issue with my running environment

To do: read and handson Understanding Variational Autoencoders (VAEs)

To do: reand and handson Patsy: Build Powerful Features with Arbitrary Python Code

Thursday 10/7

Deep learning seems unstoppable! I’m particularly impressed by the recent progress of deep learning on tabular data. This new survey paper provides an overview of the SOTA deep learning methods on tabular data. A great read for students and practitioners.

Deep Neural Networks and Tabular Data: A Survey - arxiv 2110.01889

Week 42 - October 21

Wednesday 10/20

Using fastai forums to get inspirational content for VAE with tabular data. This one sounds good: Adversarial Autoencoders (with Pytorch). And this talk demystifying bayesian stuff.

“Most of human and animal learning is unsupervised learning. If intelligence was a cake, unsupervised learning would be the cake [base], supervised learning would be the icing on the cake, and reinforcement learning would be the cherry on the cake. We know how to make the icing and the cherry, but we don’t know how to make the cake.”

Yann LeCunn

And Intuitively Understanding Variational Autoencoders mentioned by Jeremy Howard