Dumrauf, Dominic: On the hardness of computing local optima. 2011
Inhalt
- Introduction
- Optimization Problems, Intractability, and Approximation
- Local Search
- The Concept of Local Search
- Successful Applications of Local Search
- A Framework to Investigate the Complexity of Local Search
- Local Search and Game Theory: Optimization in Competition
- The Class of Congestion Games
- Central Questions of this Thesis
- Superior Questions
- Restricted Network Congestion Games
- Maximum Constraint Assignment
- Weighted Standard Set Problems
- Publications
- Roadmap of this Thesis
- Notation
- Related Work
- Early Results
- Circuit/Flip
- Traveling Salesman Problem
- MaxCut
- Local Search Versions of Satisfiability Problems
- Weighted Set Problems and Graph Problems
- Recent Results and the Connection to Game Theory
- Our Contribution
- Computing Nash Equilibria in Two-Player Restricted Network Congestion Games
- On the PLS-Complexity of Maximum Constraint Assignment
- On the Complexity of Local Search for Weighted Standard Set Problems
- A Note on the Presentation of Our Reductions
- Computing Nash Equilibria in Two-Player Restricted Network Congestion Games
- The Complexity of (2)-RUNCG and (2)-RDNCG
- The Network and the Reduction in a Nutshell
- The Reduction
- Proving the Correctness and Tightness of the Reduction
- Conclusion and a Discussion of Questions 4 and 5
- On the PLS-Complexity of Maximum Constraint Assignment
- On the Relation of Maximum Constraint Assignment to Minimum Constraint Assignment
- The General Method for the Intractability Proofs of (3,2,3)-MCA3-par and (2,3,6)-MCA2-par
- The Setting
- The Idea in a Nutshell
- Assumptions and Notation for Circuit/Flip
- The Concept of Propagation Trees
- (3,2,3)-MCA3-par is Tight PLS-Complete
- The Set of Constraints
- The Set of Variables
- The Constraint-Graph of Our Reduction
- A More Detailed Overview of the Reduction
- The Set of Predicates
- Proving the Correctness and Tightness of the Reduction
- (2,3,6)-MCA2-par is Tight PLS-Complete
- The Set of Constraints
- The Set of Variables
- Similarities and Differences to our Reduction in Section 6.3
- The Modified Constraint-Graph of Our Reduction
- The Set of Modified Predicates
- Proving the Correctness and Tightness of the Reduction
- A Reduction to Binary Logic
- Conclusion and a Discussion of Question 6
- On the Complexity of Local Search for Weighted Standard Set Problems
- How to Show Intractability of Weighted Standard Set Problems
- The PLS-Complexity of Weighted Standard Set Problems
- Assumptions and Preliminaries
- On the Complexity of Weighted-3-DimensionalMatching-(p,q) and Exact-Cover-By-3-Sets-(k)
- The Exact Complexity of SetPacking-(k)
- On the Complexity of SetSplitting-(k)
- The Exact Complexity of SetCover-(k)
- On the Complexity of TestSet-(k)
- On the Complexity of SetBasis-(k)
- On the Complexity of HittingSet-(k)
- On the Complexity of IntersectionPattern-(k)
- On the Complexity of ComparativeContainment-(k)
- Proving the Tightness of the Reductions
- On the Tractability of Weighted Standard Set Problems
- Conclusion and a Discussion of Question 7
- Conclusion and Directions for Further Research
- General Conclusion and a Discussion of Questions 1–3
- Settling the PLS-Complexity of the Problems Considered in this Thesis
- A Note on the General Structure of Our PLS Reductions
- Possible Sources of Intractability in Our PLS-Complete Problems
- Open Problems
- What is There to Take Home?
- Bibliography
