Differences between revisions 126 and 128 (spanning 2 versions)

Randomized Algorithms (summer term 2014)

The course is given by Dr. Sabine Storandt. It takes place every Monday and Wednesday from 10:15am until 11:45am in the Hörsaal 03-26 in building 51.

Topics

- Las Vegas & Monte Carlo Algorithms

- Applications for Randomized Algorithms and Data Structures

- Improving Deterministic Bounds via Randomization

- Randomization in Games and AI

- Randomized Online Algorithms

- Probabilistic Method

Script

The complete script of the lecture (including some further references) is available in the course svn repository.

The final version (last change: 28.08.14) of the script can be viewed here.

Exam

The oral exam takes place at 26.08.2014.

Here is a list of example questions that are likely to be asked in the exam.

Question Time

==> Monday, 04.08. at 16:15 and Monday, 25.08. at 10:15 (Hörsaal 03-26 in building 51). Please come prepared, i.e. read through the script before and tell me explicitly what is unclear or where you need additional explanations.

Dates

1. Preliminaries

28.04. Introduction & Basic Stochastics notes
30.04. Las Vegas and Monte Carlo Algorithms: Analysis and Concentration Bounds, LV & MC for Max 3-SAT notes

2. Randomized Algorithms and Data Structures

05.05. k-Select, Approximative Median and Quick-Sort, Sorting Lower Bound notes
07.05. Sorting Lower Bounds continued, Deterministic & Randomized Skip-Listsnotes
12.05. Universal Hashingnotes
14.05. Exercise. Focus: Deterministic & Randomized Sorting notes example code
19.05. Fingerprinting and Further Hashing Applications, Introduction to Samplingnotes
21.05. VC-dimension of Set Systems & Epsilon-Nets notes
26.05. Exercise + Good Random Samples for Systems with Small VC-dimension and Epsilon-Nets notes
02.06. Good Hitting Set Approximations notes
04.06. Route Planning and Epsilon-Nets notes
16.06. Exercise. Focus: Hitting Sets for Road Networksnotes example code

3. Getting Beyond Deterministic Bounds via Randomization

18.06. The Closest Pair problemnotes
23.06. The Min Cut problemnotes
25.06. Sublinear Algorithmsnotes
30.06. Long Path problems + Exercise (Min Cut and Connected Components)notes example code connected components example code min cut

4. Randomization in AI

02.07. Exercise (Min Cut ctd. and Long Path) + The Cow Path Problem notes
07.07. Robot Navigation in Unknown Terrain: Deterministic Algorithmsnotes
09.07. Robot Navigation in Unknown Terrain: Randomized Algorithmnotes
14.07. Robot Navigation ctd., Competitive Bidding, Exercise: Cows & Robotsnotes

5. Online Algorithms

16.07. The Paging problem notes
21.07. Renting Skies and Buying a Bahn Card notes

6. Probabilistic Method

23.07. Lovasz Local Lemmanotes

-  ⇤ ← Revision 126 as of 2014-08-22 19:56:22 → 
  Size: 7602
  Editor: adpult
  Comment:
+   ← Revision 128 as of 2014-08-28 08:35:01 → ⇥
  Size: 6331
  Editor: adpult
  Comment:
-Deletions are marked like this.
+Additions are marked like this.
 Line 25:
-The final version (last change: 22.08.14) of the script can be viewed [[https://daphne.informatik.uni-freiburg.de/svn-public/RandomizedAlgorithmsSS2014/public/script.pdf|here.]] 

Please report any kind of mistakes (grammar, spelling, content) via email. Content errors corrected after 29.07.2014 will be listed on this website for easy adaption of printed scripts.

@12.08.14: S.60 With the optimal cost being equal to αM , the resulting ratio is 1 + 2/α ====> the resulting ratio is 1 + 1/(2α)

@16.08.14: S.49 the sum of 2^{i+1}^ from i=1 to j+1 is not 2(2 ^ {j+2}^ -1) but 2(2 ^ {j+2}^ -2), and in the same flavour the sum  of 2^{i+1}^ from i=1 to j is not 2(2 ^ {j+1}^ -1) but 2(2 ^ {j+1}^ -2)

@19.08.14: S.13 |A1 | ≥ k instead of |A1 | > k

@19.08.14: S.22 Lemma 2.12 |U | ≥ (n − 1) · m + 1 instead of |U | ≥ (n − 1) (m + 1)

@20.08.14: S.39 P(n) >= (1 - 2/n) * P(n - 1) anstatt P(n) <= (1 - 2/n) * P(n - 1)

@20.08.14: S.41 FastCut has to reduce the number of nodes down to (n+1)\sqrt(2) to make the success probability larger than 1/2, also the screw up probabilty missed the '1-' term at the beginning

@20.08.14: S.49 The sum in Lemma 4.1 is n^2^+2n, not n^2^.

@22.08.14: S.52 'the robot first needs d to reach y' instead of 'the robot first needs s to reach y'

@22.08.14: S.55 'W_i \qeq W and W_i \leq' W makes no sense of course, just W_i \geq W is correct, the other term needs to be deleted
+The final version (last change: 28.08.14) of the script can be viewed [[https://daphne.informatik.uni-freiburg.de/svn-public/RandomizedAlgorithmsSS2014/public/script.pdf|here.]]