Direkt zum Inhalt springen
Computer Vision Group
TUM Department of Informatics
Technical University of Munich

Technical University of Munich

Menu
Home Teaching Winter Semester 2019 Convex Optimization for Machine Learning and Computer Vision (IN2330) (2h + 2h, 6 ECTS)

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
teaching:ws2019:cvx4cv [2019/10/10 12:04]
Zhenzhang Ye
teaching:ws2019:cvx4cv [2020/05/01 16:46] (current)
Dr. Tao Wu
Line 2: Line 2:
  
 <​html>​ <​html>​
-<b> Announcement:​ <br />  <br /> +<b> Announcement:​ </b><br /> 
-<html><​font color="​red">​The first lecture ​starts ​on 21.10. There will be NO lectures on 14.10 and 16.10. ​</​font></​b>​+The retake exam is unfortunately canceled due to the Coronavirus outbreak. ​<b><​font color="​red"> ​</​font></​b>​ 
 +<br /> 
 +The location of the final exam is updated. <br /> 
 +There is no lecture on 23.12. The tutorial session on 08.01 will be replaced by the lecture.  
 +<br /> 
 +There will be no lecture/​tutorial on 09.12 and 11.12 due to NeurIPS conference. <br /> 
 +The tutorial on 30.10 will be given by Yuesong Shen 
 +<br /> 
 +The first lecture ​takes place on 21.10. There will be no lectures on 14.10 and 16.10. ​
 </​html>​ \\ </​html>​ \\
  
Line 18: Line 26:
  
 Elements in convex analysis Elements in convex analysis
-       * Convex ​sets and convex ​functions+       * Convex ​set and convex ​function
        * Existence and uniqueness of minimizers        * Existence and uniqueness of minimizers
-       ​* ​Subdifferentials +       ​* ​Subdifferential 
-       * Convex ​conjugates +       * Convex ​conjugate 
-       ​* ​Saddle point problems and duality+       ​* ​Duality theory 
 +       * Moreau envelope
   ​   ​
 Numerical methods Numerical methods
        * Gradient-based methods        * Gradient-based methods
-       * Proximal algorithmsprimal-dual hybrid gradient method, alternating direction method of multipliers +       * Proximal algorithmsprimal-dual hybrid gradient method, alternating direction method of multipliers 
-       ​* ​Convergence ​analysis+       ​* ​Unified convergence ​analysis
        * Acceleration techniques        * Acceleration techniques
- 
  
  
 Examplary applications in machine learning and computer vision include Examplary applications in machine learning and computer vision include
-       * Training of SVMs, Logistic regression +       * Logistic regression 
-       * Image reconstruction (e.g. denoising, deblurring, inpainting)+       * Training of SVMs 
 +       * Image reconstruction (e.g. denoising, inpainting, segmentation)
        * Low-rank and sparse matrix decomposition        * Low-rank and sparse matrix decomposition
-We will implement some of them in MATLAB.+We will implement some of them in Python and MATLAB.
  
 ---- ----
Line 42: Line 51:
 == Lecture == == Lecture ==
 ** Location:** Room 02.09.023\\ ** Location:** Room 02.09.023\\
-** Time and Date:** Monday 16:15 - 18:​00\\ ​  +** Time and Date:** Monday 16:15 -- 18:​00\\ ​  
 ** Start:** October 21st, 2019 \\ ** Start:** October 21st, 2019 \\
 ** Lecturer:** [[:​members:​wuta|Dr. Tao Wu]]\\ ** Lecturer:** [[:​members:​wuta|Dr. Tao Wu]]\\
Line 49: Line 58:
 == Exercise == == Exercise ==
 ** Location:** Room 02.09.023\\ ​ ** Location:** Room 02.09.023\\ ​
-** Time and Date:** Wednesday 12:15 - 14:​00\\ ​  +** Time and Date:** Wednesday 12:15 -- 14:​00\\ ​  
 ** Start:** October 23rd, 2019\\ ** Start:** October 23rd, 2019\\
 ** Organization:​** [[:​members:​yez|Zhenzhang Ye]]\\ ​ ** Organization:​** [[:​members:​yez|Zhenzhang Ye]]\\ ​
Line 67: Line 76:
 1. get at least 75% grades of all exercises totally, i.e. sum of the grades you get in all exercises divided by the grades of all exercises (without bonus) should be >= 0.75\\ 1. get at least 75% grades of all exercises totally, i.e. sum of the grades you get in all exercises divided by the grades of all exercises (without bonus) should be >= 0.75\\
 2. present your theoretical solution during tutorial at least once in this semester.\\ 2. present your theoretical solution during tutorial at least once in this semester.\\
-You cannot improve either 1.0 or >4.0.\\+You cannot improve either 1.0 or 5.0.\\
  
 == Exam == == Exam ==
-tba+Date: February 28th, 17:00 - 18:30. \\ 
 +Place: MW 0350, Egbert-von-Hoyer-Hörsaal (5503.EG.350). \\ 
 +The final exam will be written. You are allowed to bring one A4-page, handwritten cheat sheet to the exam.
  
 +== Retake Exam ==
 +The retake exam is canceled due to Coronavirus. ​
  
-== Lecture ​Material ​== +== Lecture ​Materials ​== 
-Course ​material ​(slides and exercise sheets) can be accessed [[teaching:​ws2019:​cvx4cv:​materials|here]]. \\ \\ Send us an email if you need the password.+Course ​materials ​(slides and exercise sheets/solutions) can be accessed [[teaching:​ws2019:​cvx4cv:​materials|here]]. \\ \\ Send us an email if you need the password.

Rechte Seite

Informatik IX
Chair of Computer Vision & Artificial Intelligence

Boltzmannstrasse 3
85748 Garching

info@vision.in.tum.de