Deep Learning

Fall Semester 2017

Deep learning is an area within machine learning that deals with algorithms and models that automatically induce multi-level data representations.

In recent years, deep learning and deep networks have significantly improved the state-of-the-art in many application domains such as computer vision, speech recognition, and natural language processing. This class will cover the fundamentals of deep learning and provide a rich set of hands-on tasks and practical projects to familiarize students with this emerging technology.

Course Catalogue Info

News

11/08/2017 The webpage for this year's course is online.
01/10/2017 Uploaded slides lecture 2.
03/10/2017 Added exercise 3.
06/10/2017 Added slides lecture 3.
09/10/2017 Added exercise 4 + solutions exercise 3.
16/10/2017 Added slides lecture 4.
17/10/2017 Added exercise 5 + solutions exercise 4.
23/10/2017 Added slides lecture 5 + solutions exercise 5.
26/10/2017 Added exercise 6.
30/10/2017 Re-uploaded slides lecture 6 with added material (slide 21 & 22).
30/10/2017 Added exercise 7 + solutions exercise 6.
06/11/2017 Added slides lecture 7.
12/11/2017 Added slides lecture 8 + modified slides lecture 7.
13/11/2017 Added exercise 8.
20/11/2017 Added slides lecture 9 + exercise 9.
26/11/2017 Added slides lecture 10.
27/11/2017 Added solutions for exercise 9.
04/12/2017 Added slides lecture 11.

Course Overview

WeekDateTopicSlidesExercisesMaterial
1 25.09.2017 Introduction Tensorflow session 1 exercise 1solution
1 25.09.2017 Introduction compositional model slides 1 exercise 2 solutions 2
2 02.10.2017 Approximation theory, Rectification & Sigmoid networks slides 2 exercise 3 solutions 3
3 09.10.2017 Feedforward Networks slides 3 exercise 4 solutions 4
4 16.10.2017 Backpropagation slides 4 exercise 5 solutions 5
5 23.10.2017 Optimization slides 5 exercise 6 solutions 6
6 30.10.2017 Optimization 2 slides 6 exercise 7 solutions 7
7 06.11.2017 CNN slides 7 exercise 8 TwoLayerCNN.py
train_and_test_MNIST.py
solution 8 TwoLayerCNN_solution.py
train_and_test_MNIST_solution.py
8 13.11.2017 NLP slides 8 exercise 9
train.py
data_helper.py
text_cnn.py
glove.6B.50d.tar.gz
twitter-datasets.tar.gz
solutions 9
rnn_by_hand.py
text_cnn_solution.py
text_rnn_solution.py
9 20.11.2017 Memory & Attention slides 9 exercise 10
train.py
solutions 10
10 27.11.2017 Autoencoders slides 10
11 04.12.2017 Density Estimation & Generative Adversarial Models slides 11 exercise 11
mnist-autoencoder-template.py
solutions 11
mnist-autoencoder.py
mnist-autoencoder.ipynb

Past exams

Exam 2016

Contact

To ask a questions, please do not send emails but ask on our forum at piazza. The lecture is 263-3210-00L, you can sign up here. Please post questions there, so others can see them and share in the discussion.

Lecturer Prof. Thomas Hofmann
Head assistant Dr. Aurelien Lucchi
Assistant Dr. Javier Montoya
Assistant Gary Becigneul
Assistant Hadi Daneshmand
Assistant Gideon Dresdner
Assistant Octavian Ganea
Assistant Paulina Grnarova
Assistant Mikhail Karasikov
Assistant Yannic Kilcher
Assistant Jonas Kohler
Assistant Andreas Marfurt
Assistant Kevin Roth
Assistant Florian Schmidt

Literature

Deep Learning, An MIT Press book, Ian Goodfellow, Yoshua Bengio and Aaron Courville