Laboratory, Deep Learning Lab

Welcome to the Deep Learning Lab a joint teaching effort of the Robotics (R), Computer Vision (CV), Machine Learning (ML) and Neurorobotics (NR) Labs.
Deep learning has brought a revolution to AI research. A good understanding of the principles of deep networks and experience in training them has become one of the main assets for successful research and development of new technology in machine learning, computer vision, and robotics. In this course, we want to teach students the practical knowledge that is needed to do research with deep learning and reinforcement learning. The course is divided into two main tracks that focus on different aspects of deep learning research.

Track 1: Robotics and Reinforcement Learning (R/NR). 11LE13P-7302 and 11LE13P-7320.
Track 2: Computer Vision and Automated Machine Learning (CV/ML). 11LE13P-7305, 11LE13P-530-15 and 11LE13P-7312.

Lecture/Exercises: Tuesday, 14.00-16.00 Room: Kinohörsaal/computer pools, Building 082 Beginning: Tuesday, October 16, 2018 Communications: Join our Slack channel for updates on the course (dl-lab-freiburg.slack.com) You can join with this invitation link. Requirements: Fundamental programming skills in Python. Basic knowledge about machine learning techniques (supervised learning, classification, regression), e.g. passed the Deep Learning lecture. Some experience with the Linux toolchain (text editor, compiler, linker, debugger) is recommended. Schedule: Phase I: Neural Networks Recap 16.10.2018: Lecture 1: recap neural network basics, hand out Exercise 1 23.10.2018: Meeting solving open questions, final date to switch Tracks (done via email, or best directly via HISinOne) 30.10.2018: Lecture 2: convolutional neural networks, hand out Exercise 2, submission Exercise 1 06.11.2018: Meeting solving open questions Phase II: Split up in 2 Tracks (R/NR) and (CV/ML) 13.11.2018: Lecture 3: Track1: Imitation Learning, Track2: Semantic Segmentation, hand out Exercise 3, submission Exercise 2 20.11.2018: Meeting solving open questions 27.11.2018: Meeting solving open questions 04.12.2018: Lecture 4: Track1: Reinforcement Learning, Track2: Automated ML, hand out Exercise 4, submission Exercise 3 11.12.2018: Meeting solving open questions 18.12.2018: Presentation topics for Final Project Phase III: Final Project 08.01.2018: Start project work, submission Exercise 4 15.01.2018: Meeting solving open questions 22.01.2018: Meeting solving open questions 29.01.2018: Meeting solving open questions 03.02.2018: submission Final Project (Poster + Code) 05.02.2018: Final project poster session in Kinohörsaal from 14.00-16.00 Contact: Track 1: eitel@cs.uni-freiburg.de; valada@cs.uni-freiburg.de; hueglem@cs.uni-freiburg.de Track 2: kleinaa@cs.uni-freiburg.de; gargiani@cs.uni-freiburg.de; oliveira@cs.uni-freiburg.de; saikiat@cs.uni-freiburg.de Slides: Lecture 1: Introduction Lecture 2: CS231 CNNs Intro Hyperparameter Optimization Lecture 3 Track 1: Imitation Learning ,Track2: Semantic Segmentation Lecture 4 Track 1: RL Basics, Value-based DQN Track 2: Auto ML Final Project Track 1 Final Project Track 2 CV Final Project Track 2 ML Assignments: You can find the assignments on github Further Materal: The Stanford CS231 deep learning course A great reinforcement learning course by David Silver Berkeley RL course by Sergey Levine Deep learning book by Yoshua Bengio, Ian Goodfellow and Aaron Courville, Deep learning course by Nando de Freitas,