Plans for the week of March 16-20

Dear all, welcome back to FYS5429/9429.?

Since we have an upcoming deadline for project 1, we would like to remind of the following:

1) If you plan to write two projects, the deadline for project 1 is Friday March 20 at midnight. Please upload via canvas. You can upload your report as a PDF file or link to your GitHub/Gitlab or similar repos. These repos should contain a folder with your report, another folder with codes and an eventual folder with selected test runs. Please do include a README file which explains the tructure of the repo. If you submit a PDF file of the report, this should contain a link to your repository.

2) If you plan to write one project only, with deadline June 1, we want you all to upload a status report. This does not need to be long, 1-2 pages and should contain the status of your work, what you are planning to do, what you have achieved, if you foresee problems etc etc.?

Please upload this status report by the deadline March 20 at the same canvas link as of project 1. We will try to give you feedback on your status.

Please do not hesitate to reach out to us in case something is unclear.

Else, the plans this week are

Plans for the week March 16-20

Discussion of Autoencoders (AEs) and reminder from last week

Finalizing discussion of Linear autoencoders and the full connection to PCA

Nonlinear and regularized autoencoders and discussion of? AE codes?

Our emphasis is on:

Linear algebra and optimization,

Geometry of latent representations,

Reconstruction error minimization,

Reading recommendations

Goodfellow et al chapter 14.

Rashcka et al. Their chapter 17 contains a brief introduction only.

Deep Learning Tutorial on AEs from Stanford University at?http://ufldl.stanford.edu/tutorial/unsupervised/Autoencoders/

Building AEs in Keras at?https://blog.keras.io/building-autoencoders-in-keras.html

Introduction to AEs in TensorFlow at?https://www.tensorflow.org/tutorials/generative/autoencoder

Grosse, University of Toronto, Lecture on AEs at?http://www.cs.toronto.edu/~rgrosse/courses/csc321_2017/slides/lec20.pdf

Bank et al on AEs at?https://arxiv.org/abs/2003.05991

Baldi and Hornik, Neural networks and principal component analysis: Learning from examples without local minima, Neural Networks 2, 53 (1989)

The lecture notes with codes etc are at https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/pub/week9/ipynb/week9.ipynb

Best wishes to you all,

Morten, Oda and Ruben