Lecture: Deep Learning for Visual Recognition
Course
- Lecturer(s):
- Start: 18.10.2018
- Dates: Thu. 10:15 - 11:45, CP1-HSZ / Hörsaal 3
- Course number: MA-INF 2313
- Curriculum: Master
- Effort: 6 CP
- Exams: 20 February 2019, 9:00-12:00, HS 5+6, 7 and 27 March 2019, 15:00-18:00 HS 5+6, 7
Exercises
- Tutor(s):
- Start: 25.10.2018
- Dates: Thu. 14:00 - 16:00 U1.042
Description
Neural networks are making a comeback! Deep learning has taken over the machine learning community by storm, with success both in research and commercially. Deep learning is applicable over a range of fields such as computer vision, speech recognition, natural language processing, robotics, etc. This course will introduce the fundamentals of neural networks and then progress to state-of-the-art convolutional and recurrent neural networks as well as their use in applications for visual recognition. Students will get a chance to learn how to implement and train their own network for visual recognition tasks such as object recognition, image segmentation and caption generation. No formal pre-requisites. Students should already be comfortable with concepts in probability theory and optimization and are recommended to have taken at least one course in machine learning or computer vision. Exercises will be a mix of theory and practical (Python).
News
First lecture starts on October 18, 2018. See you there!
Please inscribe yourself into the mailinglist at: 
https://lists.iai.uni-bonn.de/mailman/listinfo.cgi/vl-dl
In case you have a problem understanding something, questions related to exercises/projects, please always feel free to write to the mailing list. This should be a place where you students can talk freely about the lecture, so please do not hesitate to ask and reply!
** UPDATED INFO : Regarding the exercise, you can either work by yourself or in groups of 3. The submitted solutions should always include your (and your partner's) name and matr. number ! The theoretical solutions can be submitted via email to the tutor or directly in person during the exercise session. The programming assignment needs to be presented in person during the exercise session. If you cannot make it because of time conflicts, you can ask the tutor for an appointment and then present your solution. **
** Please email the tutor with your preferred project along with the name of your group partners latest by Tue 11.12.2018 **
** A cheat sheet for accessing and using the GPU Server is now available ! **
Additional Documents
- DL2018 CourseInfo (PDF document, 52 KB)
- DL GPU Access (PDF document, 71 KB)
- DL Projects (PDF document, 1.8 MB)
- Gradient Descent and Backpropagation Example (PDF document, 1.0 MB)
Slides
- 01 Intro 181018 (PDF document, 41 MB)
- 02 MLbasics 251018 (PDF document, 24 MB)
- 03 NNintro 081118 (PDF document, 1.9 MB)
- 04 NNintro2 151118 (PDF document, 1.9 MB)
- 05 DeepOpt 221118 (PDF document, 2.5 MB)
- 06 Regularization 061218 (PDF document, 2.0 MB)
- 07 CNN (PDF document, 5.2 MB)
- 08 CNNII (PDF document, 29 MB)
- 09 RNNI 100119 (PDF document, 2.3 MB)
- 10 RNNII 170119 (PDF document, 4.9 MB)
- 11 Autoencoders 240119 (PDF document, 2.5 MB)
- 12 GenerativeI (PDF document, 11.3 MB)
Assignment Sheets
|
|
Exercise 1: MLBasics Assignment sheet (PDF document, 170 KB)
|
|
|
Exercise 2: NeuralNetworks Assignment sheet (PDF document, 108 KB)
|
|
|
Exercise 3: OptimizationRegularization Assignment sheet (PDF document, 119 KB)
|
|
|
Exercise 4: CNNs Assignment sheet (PDF document, 135 KB)
|
|
|
Exercise 5: RNNs Assignment sheet (PDF document, 254 KB)
|
|
|
Exercise 6: Autoencoders Assignment sheet (PDF document, 144 KB) |