Deep Learning

Fall Semester 2016

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

11/08/2016 The webpage for this year's course is online.
27/09/2016 Added slides lecture 1 + exercise 1.
03/10/2016 Added slides lecture 2 + exercise 2
10/10/2016 Added slides lecture 3.
10/10/2016 Added exercise 3 + solutions for exercise 2.
17/10/2016 Added slides lecture 4.
18/10/2016 Added exercise 4 + solutions for exercise 3.
24/10/2016 Added slides lecture 5.
25/10/2016 Added exercise 5 + solutions for exercise 4.
01/11/2016 Added exercise 6 + solutions for exercise 5.
08/11/2016 Updated slides lecture 6 + added exercises 7 & solutions for exercise 6.
14/11/2016 Added slides lecture 7 + exercise 8.
21/11/2016 Added slides lecture 8.
22/11/2016 Added exercise 9.
28/11/2016 Updated lecture 8.
18/01/2017 Added sample questions for the exam.

Course Overview

WeekDateTopicSlidesExercisesMaterial
1 26.09.2016 Feedforward networks: architecture slides 1 exercise 1
2 03.10.2016 Feedforward networks: learning slides 2 exercise 2 solutions 2
3 10.10.2016 Optimization for neural networks slides 3 exercise 3 solutions 3
4 17.10.2016 Regularization for neural networks slides 4 exercise 4 solutions 4
5 24.10.2016 Convolutional Neural Networks slides 5 exercise 5 solutions 5
6 31.10.2016 Recurrent and recursive neural networks slides 6 exercise 6
TwoLayerCNN.py
train_and_test_MNIST.py
solutions 6
code solution
7 07.11.2016 Recurrent and recursive neural networks
(part 2)
exercise 7 solutions 7
code solution
8 14.11.2016 Factor models, autoencoders, latent representations slides 7 exercise 8 code solution
9 21.11.2016 - 28.11.2016 Undirected Deep Models
& Generative models
slides 8 exercise 9
Template (.ipynb)
Template (.py)
solutions 9
- 05.12.2016 No lecture (project) No exercise session
- 12.12.2016 No lecture (project) No exercise session
- 18.01.2017 - sample exam questions

Contact

Lecturer Prof. Thomas Hofmann
Assistant Aurelien Lucchi
Assistant Florian Schmidt
Assistant Yannic Kilcher
Assistant Gary Becigneul
Assistant Javier Montoya

Literature

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