index.md

Deep Learning (CAS machine intelligence, 2020)

This course in deep learning focuses on practical aspects of deep learning. We therefore provide jupyter notebooks (complete overview of all notebooks used in the course).

For doing the hands-on part on your own computer you can either install anaconda (details and installation instruction) or use the provided a docker container (details and installation instruction).

To easily follow the course please make sure that you are familiar with the some basic math and python skills.

Info for the projects

You can join together in small groups and choose a topic for your DL project. You should prepare a poster and a spotlight talk (5 minutes) which you will present on the last course day. To get some hints how to create a good poster you can check out the links that are provided in poster_guidelines.pdf

If you need free GPU resources, we might want to follow the instructions how to use google colab. Help for preparing a hdf5 file from your images you can be found in the example Notebook.

Examples for projects from the DL course 2018 and 2019 can be found here.

Fill in the Title and the Topic of your Projects until 24.03.2020 here

Other resources

We took inspiration (and sometimes slides / figures) from the following resources.

• Probabilistic Deep Learning (DL-Book) Probabilistic Deep Learning. This book is by the tensorchiefs and covers the increasingly popular probabilistic approach to deep learning .

• Deep Learning Book (DL-Book) http://www.deeplearningbook.org/. This is a quite comprehensive book which goes far beyond the scope of this course.

• Convolutional Neural Networks for Visual Recognition http://cs231n.stanford.edu/, has additional material and youtube videos of the lectures. While the focus is on computer vision, it also treats other topics such as optimization, backpropagation and RNNs. Lecture notes can be found at http://cs231n.github.io/.

• More TensorFlow examples can be found at dl_tutorial

• Another applied course in DL: TensorFlow and Deep Learning without a PhD

Dates

The course is split in 8 sessions, each 4 lectures long.

Day	Date	Time
1	18.02.2020	13:30-17:00
2	25.02.2020	13:30-17:00
3	03.03.2020	13:30-17:00
4	10.03.2020	13:30-17:00
5	17.03.2020	13:30-17:00
6	24.03.2020	09:00-12:30
7	31.03.2020	13:30-17:00
8	07.04.2020	09:00-12:30

Syllabus (might change during course)

Day	Topic and Slides	Additional Material	Exercises and homework
1	Introduction, Fully Connected Networks, Keras slides	Network Playground	01_simple_forward_pass 02_fcnn_with_banknote
2	Looking back at fcNN, working with loss curves, convolutional neural networks slides	Understanding convolution	03_fcnn_mnist 04_fcnn_mnist_shuffled 05_cnn_edge_lover 06_cnn_mnist_shuffled 07_cifar10_norm
3	Tricks of the trade in CNNs slides	Understanding CNNs	08_cifar10_tricks 09_1DConv
4	Details: Backpropagation in DL, MaxLike-Principle slides	TODO	10_linreg_tensorflow 11_backpropagation maxlik
no lecture due to corona crisis	no lecture due to corona crisis	no lecture due to corona crisis	no lecture due to corona crisis
6	Probabilistic Models slides_part_1 slides_part_2		13_linreg_with_tfp 14_poisreg_with_tfp
7	Probabilistic models in the wild slides_part_1 slides_part_2		15_zipreg_with_tfp 16_linreg_with_tfp_const_sigma 17_faces_regression 18_elephant_in_the_room
8	Bayesian Deep Learning slides_part_1 slides_part_2	TODO	20_cifar10_classification_mc_and_vi

Tensorchiefs are Oliver Dürr, Beate Sick and Elvis Murina.