Project proposals for 236331

Previous update: April 29, 2001
Last update: April, 30, 2001
Back to Course homepage.
Below you find several project descriptions suitable for the final project of the course. You should contact Julian Marino in room 619 (TAUB, CS), Tel: 829 4873. You can fix appointments by phone or e-mail (julian@cs.technion.ac.il).
Projects need approval and several meetings with us are compulsory part of projects

    Definability/Complexity Catalogue

    In these projects the task is to collect material about one of the following problems, and discuss its complexity and definability status.

  1. Reachability (in graphs).
    Relevant literature:
  2. Planarity of graphs.
    Relevant literature
  3. Colorability of graphs
    Relevant literature
  4. Hamiltonicity of graphs
    Relevant literature
  5. Matchings
    Relevant literature
  6. Eulerian circuits
    Relevant literature
  7. Satsifiability of propositional logic (coded as a graph)
    Relevant literature
  8. The problem of your choice (after approval of J. Marino or J. Makowsky)
In each of the following projects, the goal would be to select material and prepare it for a 2-3 hour presentation in the style of the course. (Maximally material for 20 slides).

    Picture languages

    Picture languages are sets of rectangular coloured (n x n) grids (like pixels on an (n x n)-screen). There are several projects here:

  1. Find examples of picture languages which are of particular interest. Classify them by complexity (in P, NP, worse, better), by definability (in FOL, MSOL, ExistMSOL, SOL).
    A specially amusing examples comes from the MINESWEEPER game, and is NP-complete. For futher reference you have to go to the new CS-LIBRARY.
    Thanks to Yaron Welner for suggesting this project.
  2. Characterize the ExistMSOL picture languages via tiling systems.
    Relevant paper are
  3. Discuss the quantifier hierarchy of MSOL for pictures.
    Relevant papers are definable picture languages.

    Quantifier prefixes of Second Order Horn formulas

    In the course we showed that truth in structures of existential second order Horn formulas can be done in polynomial time. This is true for arbitrary second order Horn formulas. In fact, every second order Horn formula is equivalent to an existential one. The proof of this is the content of this project.

    P-reductions vs transducers in SOL Horn Logic.

    See lecture from May 3, 2001

Project assignment

  1. Yaron Welner, Examples of Picture languages including MINESWEEPER. Complexity and Definability of each example.
    final received: 18.11.01 grade 93
  2. Alexander Glikson Tree automata and Monadic Second Order Logic
    final received: 5.11.01 Grade 98
  3. Jonathan Ishai and Benjamin Sznajder Quantifier prefixes of Second Order Horn formulas
    final received: 04.12.01 Grade: Ishai 80, Sznajder 85
  4. Miriam Wallach SAT and its definability and complexity
    final received: 22.10.01 Grade 85
  5. Arieh Bibliowicz and Eitan Yoshuah Datacompression and definability, after Afrati, Leiss and de Rougemont ps-file
    final received: 04.12.01 Grade for both 90
  6. Yonit Magid Reachability in graphs
    final received: 5.11.01 Grade 95
  7. Gilad Ben-Avi Characterize the ExistMSOL picture languages via tiling systems
    final received: 14.12.01 Grade 90
  8. Ron Gross Connectivity and Separability in graphs
    final received: 10.10.01 Grade 90