University COVID-19 update

The University of Waterloo is constantly updating our most Frequently Asked Questions.

Questions about buildings and services? Visit the list of Modified Services.

Please note: The University of Waterloo is closed for all events until further notice.

Tutte seminar - Mike MolloyExport this event to calendar

Friday, November 30, 2007 — 3:30 PM to 4:30 PM EST

The Resolution Complexity of Random Constraint Satisfaction Problems

Speaker: Mike Molloy
Affiliation: University of Toronto
Room: Mathematics & Computer Building (MC) 5158

Abstract:

Constraint satisfaction problems (CSP's) are generalizations of CNF boolean formulas. We are given a set of variables, and a domain of values that the variables can take. Some subsets of the variables have "constraints" which restrict the value-assignments that can be made to those variables. A problem is satisfiable if there is an assignment of values to the variables which does not violate any of the constraints. Some well-studied special cases are k-SAT, graph colourability and graph homomorphism. 
Random instances of various CSP's appear to be very difficult to solve in practice. Chvátal and Szemerédi proved a theorem explaining this for the case of random 3-SAT: with high probability, the resolution complexity of a random 3-SAT formula with a linear number of clauses will be exponentially high. This implies an exponential lower bound on the running time of any resolution-based algorithm (such as Davis-Putnam and its many variants).

In this talk, we discuss a wide class of models of random constraint satisfaction problems. The main result is that we determine which models from this class will, with high probability, have exponentially high resolution complexity for any linear number of constraints.

This is joint work with Siu On Chan.

Location 
MC - Mathematics & Computer Building
5158
200 University Avenue West

Waterloo, ON N2L 3G1
Canada

S M T W T F S
30
31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1
2
3
  1. 2020 (92)
    1. November (1)
    2. October (1)
    3. September (12)
    4. August (11)
    5. July (17)
    6. June (11)
    7. May (6)
    8. March (11)
    9. February (11)
    10. January (11)
  2. 2019 (167)
    1. December (5)
    2. November (15)
    3. October (18)
    4. September (15)
    5. August (9)
    6. July (17)
    7. June (18)
    8. May (16)
    9. April (9)
    10. March (24)
    11. February (13)
    12. January (8)
  3. 2018 (138)
  4. 2017 (103)
  5. 2016 (137)
  6. 2015 (136)
  7. 2014 (88)
  8. 2013 (48)
  9. 2012 (39)
  10. 2011 (36)
  11. 2010 (40)
  12. 2009 (40)
  13. 2008 (39)
  14. 2007 (15)