SlideShare a Scribd company logo
1 of 3
MIDTERM REVIEW SESSION FOR CS261



                                         1. What's this about?

   approximation algorithm

       •   in theory: a rigorous way to deal with NP-hardness. focus on worst-case
           guarantee.
                in practice: even a 3/4-approximation is unsatisfactory. cutting plane
                 methods. integer program solver. local search. heuristics.
       •   common: both often use optimal LP value as a benchmark.
       •   there's always a gap between theory and practice.


                                   2. Combinatorial Approach

       Vertex Cover:             unnatural greedy is 2-approx.              size of max matching as a
           lower bound.
       Steiner Tree:            non-metric version reducible to the metric version.                then do
           (1) double the tree (dfs) (2) traverse (3) shortcut (justied by metric ineq.).
       TSP:      general cost + non-repetition is not O(1)-approximable.                        if you can
           repeat, then metric is ne.
             •   2-approx: double the MST + shortcut.
             •   3/2-approx: MST + min-cost perfect matching on odd-degree nodes
                 (at cost=half of optimal) + eulerian cycle + shortcut.
       Set Cover: has VC as a special case.
           L.B.: if we have a subset D of elements,                        and every set   Si   contains at
                 most   t   elements of      D,   then opt is at least      |D|/t.
             •   when the         i-th   element   xi   is covered,   ci   new elements were covered.
                                                             1
                 amortize the cost: cost(xi )           =    ci
             •   opt=       i   cost(xi ).
             •   every set covers at most           ci of the n − i + 1     elements. So we have a l.b.
                                         n−i+1
                 on opt:    opt ≥          ci (n   − i + 1) · cost(xi )

                                         3. Linear Programming

3.1.   LP.
       primal: max cT x subject to Ax ≤ b, x ≥ 0. (called packing LP if A ≥ 0)
       dual: min bT y subject to AT y ≥ c, y ≥ 0. (called covering LP if A ≥ 0)
             (more often primal is minimization problem, then dual is the maximiza-
           tion problem.)
       outcome: feasible. infeasible. unbounded.
       meaning: highest lower-bound on primal optimal.
       weak duality: primal opt ≥ dual opt
       strong duality: primal opt = dual opt, except when both infeasible.
       duality in form: inequality ↔ nonnegativity, equality ↔ unconstrained.
       complexity: solvable in PTIME. even simplex does well.
                                                         1
MIDTERM REVIEW SESSION FOR CS261                                        2


3.2.   ILP and LP Relaxations.
        ILP: linear constraints + integer           variables. often NP-hard. (still often fea-
          sible in practice. using cplex)
       LP Relaxation: simply replace integrality constraint to like x ≥ 0.
       What's A Formulation: It's kinda fuzzy. But when you use it, it needs
          ideally: feasible solution to ILP ↔ feasible solution to problem
          also work: optimal solution to ILP ↔ optimal solution to problem
          bottomline: the LP optimal value should give a bound on optimal                         to
                the problem.(and hopefully not too far away)
       Lower Bounds:
            •   the optimal value of LP relaxation fractional optimal
            •   objective value of any feasible dual solution

3.3.   Weighted Vertex Cover.
       LP Rounding: solve LP. pick the vertex i xv ≥ 1/2.                    (1. ensures it gives a
          VC, 2. lose by a factor of 2).
       Primal Dual:      dual   yu,v .
            •   old alg: if (u,v) uncovered, set xu , xv , yu,v = 1. matching cast as a
                           y(u,v) = 1 if (u,v) in M.
                dual sol: set
            • use twice of dual y(u,v) to construct payments pv for costs of vertices,
              while preserving dual feasibility. pick v if fully-paid for.

3.4.   Weighted Set Cover.
       LP Rounding: we round             w.p.      x∗
                                                    i   independently, may not be a cover. but
          each element with probability at least 1-1/e covered.
            x:   repeat until all elements covered.               after log iterations, still exists
                uncovered elements with exponentially small probability.
       Primal Dual:      eective cost of a set          cj =        wi
                                                                newlycovered . same logic as before.
          in fact dual   pj =eective      cost.
            •   in fact, the primal is fully-paid for by eective payments, scaled down
                                            1
                by   H(m)   is feasible:
                                           H(m)         j∈S   pj ≤ c(S).

                                 4. Max Flow / Min Cut

4.1.   Basic Concepts.
       input: directed graph with capacities c : E → [0, ∞) and s, t pair.
       ow: f : E → [0, ∞) s.t. (1) upper-bounded by capacity, and (2)                      satises
          conservation constraints.

size of a ow: sum of    f from source. convention: ow/capacity
       cut: partition (A, V A) s.t. s ∈ A, t ∈ A. c(A) = u∈A,v∈A,(u,v)∈E c(u, v).
                                              /                    /
       inequalities: for any s-t cut A: cost(f ) = v∈V :(s,v)∈E f (s, v) = f (A) ≤
            a∈A,b∈A,(a,b)∈E f (u, v) ≤
                 /                      a∈A,b∈A,(a,b)∈E c(u, v) = c(A).
                                              /
       witness of optimality: a cut A that's saturated.
4.2.   Ford-Fulkerson.
       residual network: capacity = how much more ow you can push.
       augmenting path: s-t path with positive capacity arcs.
       Ford-Fulkerson: while an augmenting path can be found, push the                        maxi-
          mum amount of ow on the path.
MIDTERM REVIEW SESSION FOR CS261                                          3


Termination:           let A be set of nodes to which you can have augmenting path
    to from     s.   the cut across    A   is saturated:   f (a, b) = c(a, b).   and   f (b, a) = 0.
Implication: Max-Flow = Min-Cut
time complexity: as bad as |C| for bad choices of aug paths.
                             5. On Writing Solutions

•   Common Main ingredients:
         Formulation, Justication, Algorithm, Runtime, Feasibility, Approxi-
          mation

Formulation: make all quanties explicit
Justication: you need to show that LP optimal gives a bound on optimal.
Algorithm: Explicitly state your algorithm.
Runtime: State it. If you need to solve LP exactly, mention that this can be
    done in PTIME.
Feasibility: Show it.
Approximation: Be explicit about all important quantities, and the inequal-
    ities. be explict if you use weak duality.

•   Miscellaneous Stu
         taking dual is a mechanical process. don't be too smart.
         (optional) please don't use matrix notations when you are asked to
          write LPs.
         be explicit about whether you round it independently or dependently.
          be explicit about application of linearity of expectation.
         use   xv , yu,v   etc. not   ei,j .   no reuse of symbols.
         randomized rounding is a general method. does not refer to any par-
          ticular rounding algorithm.

More Related Content

What's hot

A Machine-Assisted Proof of Gödel's Incompleteness Theorems
A Machine-Assisted Proof of Gödel's Incompleteness TheoremsA Machine-Assisted Proof of Gödel's Incompleteness Theorems
A Machine-Assisted Proof of Gödel's Incompleteness TheoremsLawrence Paulson
 
Topological Inference via Meshing
Topological Inference via MeshingTopological Inference via Meshing
Topological Inference via MeshingDon Sheehy
 
P, NP and NP-Complete, Theory of NP-Completeness V2
P, NP and NP-Complete, Theory of NP-Completeness V2P, NP and NP-Complete, Theory of NP-Completeness V2
P, NP and NP-Complete, Theory of NP-Completeness V2S.Shayan Daneshvar
 
Functionalanalysis ejemplos infinitos
Functionalanalysis ejemplos infinitosFunctionalanalysis ejemplos infinitos
Functionalanalysis ejemplos infinitosSualín Rojas
 
Heuristics for counterexamples to the Agrawal Conjecture
Heuristics for counterexamples to the Agrawal ConjectureHeuristics for counterexamples to the Agrawal Conjecture
Heuristics for counterexamples to the Agrawal ConjectureAmshuman Hegde
 
Mit2 092 f09_lec23
Mit2 092 f09_lec23Mit2 092 f09_lec23
Mit2 092 f09_lec23Rahman Hakim
 
Mit2 092 f09_lec19
Mit2 092 f09_lec19Mit2 092 f09_lec19
Mit2 092 f09_lec19Rahman Hakim
 
Optimization
OptimizationOptimization
OptimizationSpringer
 
Nies cuny describing_finite_groups
Nies cuny describing_finite_groupsNies cuny describing_finite_groups
Nies cuny describing_finite_groupsAndre Nies
 
Machine learning session 9
Machine learning session 9Machine learning session 9
Machine learning session 9NirsandhG
 
Pushforward of Differential Forms
Pushforward of Differential FormsPushforward of Differential Forms
Pushforward of Differential FormsHeinrich Hartmann
 

What's hot (20)

A Machine-Assisted Proof of Gödel's Incompleteness Theorems
A Machine-Assisted Proof of Gödel's Incompleteness TheoremsA Machine-Assisted Proof of Gödel's Incompleteness Theorems
A Machine-Assisted Proof of Gödel's Incompleteness Theorems
 
Topological Inference via Meshing
Topological Inference via MeshingTopological Inference via Meshing
Topological Inference via Meshing
 
P, NP and NP-Complete, Theory of NP-Completeness V2
P, NP and NP-Complete, Theory of NP-Completeness V2P, NP and NP-Complete, Theory of NP-Completeness V2
P, NP and NP-Complete, Theory of NP-Completeness V2
 
Analysis of algorithms
Analysis of algorithmsAnalysis of algorithms
Analysis of algorithms
 
Functionalanalysis ejemplos infinitos
Functionalanalysis ejemplos infinitosFunctionalanalysis ejemplos infinitos
Functionalanalysis ejemplos infinitos
 
Chapter 3
Chapter 3Chapter 3
Chapter 3
 
Mark Rscribe
Mark RscribeMark Rscribe
Mark Rscribe
 
Lesson 5: Continuity
Lesson 5: ContinuityLesson 5: Continuity
Lesson 5: Continuity
 
Heuristics for counterexamples to the Agrawal Conjecture
Heuristics for counterexamples to the Agrawal ConjectureHeuristics for counterexamples to the Agrawal Conjecture
Heuristics for counterexamples to the Agrawal Conjecture
 
Mit2 092 f09_lec23
Mit2 092 f09_lec23Mit2 092 f09_lec23
Mit2 092 f09_lec23
 
CRMS Calculus 2010 May 17, 2010
CRMS Calculus 2010 May 17, 2010CRMS Calculus 2010 May 17, 2010
CRMS Calculus 2010 May 17, 2010
 
Mit2 092 f09_lec19
Mit2 092 f09_lec19Mit2 092 f09_lec19
Mit2 092 f09_lec19
 
Concept learning
Concept learningConcept learning
Concept learning
 
Optimization
OptimizationOptimization
Optimization
 
Nies cuny describing_finite_groups
Nies cuny describing_finite_groupsNies cuny describing_finite_groups
Nies cuny describing_finite_groups
 
Machine learning session 9
Machine learning session 9Machine learning session 9
Machine learning session 9
 
Pushforward of Differential Forms
Pushforward of Differential FormsPushforward of Differential Forms
Pushforward of Differential Forms
 
Iteration Techniques
Iteration TechniquesIteration Techniques
Iteration Techniques
 
Chap8 new
Chap8 newChap8 new
Chap8 new
 
01fourier
01fourier01fourier
01fourier
 

Viewers also liked

Android applications with flash
Android applications with flashAndroid applications with flash
Android applications with flashDaniel Downs
 
Opt extensive and overview of excise tax. dst and ctc.feb.2011
Opt extensive and overview of excise tax. dst and ctc.feb.2011Opt extensive and overview of excise tax. dst and ctc.feb.2011
Opt extensive and overview of excise tax. dst and ctc.feb.2011Phil Taxation
 
Individual income tax.feb.2011
Individual income tax.feb.2011Individual income tax.feb.2011
Individual income tax.feb.2011Phil Taxation
 
Corporate income tax.feb.2011
Corporate income tax.feb.2011Corporate income tax.feb.2011
Corporate income tax.feb.2011Phil Taxation
 
Allowable deductions.feb.2011
Allowable deductions.feb.2011Allowable deductions.feb.2011
Allowable deductions.feb.2011Phil Taxation
 
Intro to internet marketing brown bag workshop july 2012
Intro to internet marketing   brown bag workshop july 2012Intro to internet marketing   brown bag workshop july 2012
Intro to internet marketing brown bag workshop july 2012Calibrate Marketing & Mentoring
 
Project overview downs research
Project overview downs researchProject overview downs research
Project overview downs researchDaniel Downs
 
Designing for android tablets smashing mobile
Designing for android tablets   smashing mobileDesigning for android tablets   smashing mobile
Designing for android tablets smashing mobileDaniel Downs
 
Variables chapter 2 php book
Variables chapter 2 php bookVariables chapter 2 php book
Variables chapter 2 php bookDaniel Downs
 
R7035 phenomenon powerpoint_downs_d mod 8
R7035  phenomenon powerpoint_downs_d mod 8R7035  phenomenon powerpoint_downs_d mod 8
R7035 phenomenon powerpoint_downs_d mod 8Daniel Downs
 
Diaschau_startworkshop_NAWARO
Diaschau_startworkshop_NAWARODiaschau_startworkshop_NAWARO
Diaschau_startworkshop_NAWAROAlexander Traxler
 
Assignment slide share
Assignment  slide shareAssignment  slide share
Assignment slide shareAstigers
 
BTS 143 summer vacation
BTS 143 summer vacation BTS 143 summer vacation
BTS 143 summer vacation Astigers
 
Assignment 4 SlideShare Stigers
Assignment 4 SlideShare StigersAssignment 4 SlideShare Stigers
Assignment 4 SlideShare StigersAstigers
 
Ξεριζωμός και προσφυγιά των Ελλήνων του Πόντου
Ξεριζωμός και προσφυγιά των Ελλήνων του ΠόντουΞεριζωμός και προσφυγιά των Ελλήνων του Πόντου
Ξεριζωμός και προσφυγιά των Ελλήνων του Πόντουchristihai
 
Income tax intro topics.feb.2011
Income tax intro topics.feb.2011Income tax intro topics.feb.2011
Income tax intro topics.feb.2011Phil Taxation
 
μεταναστευση
μεταναστευσημεταναστευση
μεταναστευσηchristihai
 
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2christihai
 
General principles/Fundamentals of Taxation
General principles/Fundamentals of TaxationGeneral principles/Fundamentals of Taxation
General principles/Fundamentals of TaxationPhil Taxation
 

Viewers also liked (20)

Android applications with flash
Android applications with flashAndroid applications with flash
Android applications with flash
 
Opt extensive and overview of excise tax. dst and ctc.feb.2011
Opt extensive and overview of excise tax. dst and ctc.feb.2011Opt extensive and overview of excise tax. dst and ctc.feb.2011
Opt extensive and overview of excise tax. dst and ctc.feb.2011
 
Individual income tax.feb.2011
Individual income tax.feb.2011Individual income tax.feb.2011
Individual income tax.feb.2011
 
Corporate income tax.feb.2011
Corporate income tax.feb.2011Corporate income tax.feb.2011
Corporate income tax.feb.2011
 
Allowable deductions.feb.2011
Allowable deductions.feb.2011Allowable deductions.feb.2011
Allowable deductions.feb.2011
 
Intro to internet marketing brown bag workshop july 2012
Intro to internet marketing   brown bag workshop july 2012Intro to internet marketing   brown bag workshop july 2012
Intro to internet marketing brown bag workshop july 2012
 
Project overview downs research
Project overview downs researchProject overview downs research
Project overview downs research
 
Designing for android tablets smashing mobile
Designing for android tablets   smashing mobileDesigning for android tablets   smashing mobile
Designing for android tablets smashing mobile
 
Variables chapter 2 php book
Variables chapter 2 php bookVariables chapter 2 php book
Variables chapter 2 php book
 
R7035 phenomenon powerpoint_downs_d mod 8
R7035  phenomenon powerpoint_downs_d mod 8R7035  phenomenon powerpoint_downs_d mod 8
R7035 phenomenon powerpoint_downs_d mod 8
 
Diaschau_startworkshop_NAWARO
Diaschau_startworkshop_NAWARODiaschau_startworkshop_NAWARO
Diaschau_startworkshop_NAWARO
 
Assignment slide share
Assignment  slide shareAssignment  slide share
Assignment slide share
 
BTS 143 summer vacation
BTS 143 summer vacation BTS 143 summer vacation
BTS 143 summer vacation
 
Assignment 4 SlideShare Stigers
Assignment 4 SlideShare StigersAssignment 4 SlideShare Stigers
Assignment 4 SlideShare Stigers
 
Ξεριζωμός και προσφυγιά των Ελλήνων του Πόντου
Ξεριζωμός και προσφυγιά των Ελλήνων του ΠόντουΞεριζωμός και προσφυγιά των Ελλήνων του Πόντου
Ξεριζωμός και προσφυγιά των Ελλήνων του Πόντου
 
Income tax intro topics.feb.2011
Income tax intro topics.feb.2011Income tax intro topics.feb.2011
Income tax intro topics.feb.2011
 
μεταναστευση
μεταναστευσημεταναστευση
μεταναστευση
 
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2
μεταναστευση των ελληνων σάβιακλη φωτεινή τσότσου τάνια α΄2
 
Pub
PubPub
Pub
 
General principles/Fundamentals of Taxation
General principles/Fundamentals of TaxationGeneral principles/Fundamentals of Taxation
General principles/Fundamentals of Taxation
 

Similar to Review

Lattices, sphere packings, spherical codes
Lattices, sphere packings, spherical codesLattices, sphere packings, spherical codes
Lattices, sphere packings, spherical codeswtyru1989
 
Relaxed Utility Maximization in Complete Markets
Relaxed Utility Maximization in Complete MarketsRelaxed Utility Maximization in Complete Markets
Relaxed Utility Maximization in Complete Marketsguasoni
 
Cheatsheet recurrent-neural-networks
Cheatsheet recurrent-neural-networksCheatsheet recurrent-neural-networks
Cheatsheet recurrent-neural-networksSteve Nouri
 
Mit2 092 f09_lec21
Mit2 092 f09_lec21Mit2 092 f09_lec21
Mit2 092 f09_lec21Rahman Hakim
 
NIPS2007: learning using many examples
NIPS2007: learning using many examplesNIPS2007: learning using many examples
NIPS2007: learning using many exampleszukun
 
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdf
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdfUnit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdf
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdfyashodamb
 
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...Varad Meru
 
Linear Programming
Linear ProgrammingLinear Programming
Linear Programmingknspavan
 
Solution of nonlinear_equations
Solution of nonlinear_equationsSolution of nonlinear_equations
Solution of nonlinear_equationsTarun Gehlot
 
Linear programming in computational geometry
Linear programming in computational geometryLinear programming in computational geometry
Linear programming in computational geometryhsubhashis
 
signal and system Dirac delta functions (1)
signal and system Dirac delta functions (1)signal and system Dirac delta functions (1)
signal and system Dirac delta functions (1)iqbal ahmad
 
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8Giáo trình Phân tích và thiết kế giải thuật - CHAP 8
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8Nguyễn Công Hoàng
 
SURF 2012 Final Report(1)
SURF 2012 Final Report(1)SURF 2012 Final Report(1)
SURF 2012 Final Report(1)Eric Zhang
 
Linear programming in computational geometry
Linear programming in computational geometryLinear programming in computational geometry
Linear programming in computational geometrySubhashis Hazarika
 
Project in Calcu
Project in CalcuProject in Calcu
Project in Calcupatrickpaz
 
Design and Analysis of Algorithms Exam Help
Design and Analysis of Algorithms Exam HelpDesign and Analysis of Algorithms Exam Help
Design and Analysis of Algorithms Exam HelpProgramming Exam Help
 
Limits and continuity powerpoint
Limits and continuity powerpointLimits and continuity powerpoint
Limits and continuity powerpointcanalculus
 

Similar to Review (20)

Lattices, sphere packings, spherical codes
Lattices, sphere packings, spherical codesLattices, sphere packings, spherical codes
Lattices, sphere packings, spherical codes
 
Relaxed Utility Maximization in Complete Markets
Relaxed Utility Maximization in Complete MarketsRelaxed Utility Maximization in Complete Markets
Relaxed Utility Maximization in Complete Markets
 
Neural Networks - How do they work?
Neural Networks - How do they work?Neural Networks - How do they work?
Neural Networks - How do they work?
 
Cheatsheet recurrent-neural-networks
Cheatsheet recurrent-neural-networksCheatsheet recurrent-neural-networks
Cheatsheet recurrent-neural-networks
 
Mit2 092 f09_lec21
Mit2 092 f09_lec21Mit2 092 f09_lec21
Mit2 092 f09_lec21
 
NIPS2007: learning using many examples
NIPS2007: learning using many examplesNIPS2007: learning using many examples
NIPS2007: learning using many examples
 
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdf
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdfUnit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdf
Unit-3 greedy method, Prim's algorithm, Kruskal's algorithm.pdf
 
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passin...
 
Linear Programming
Linear ProgrammingLinear Programming
Linear Programming
 
Chemistry Assignment Help
Chemistry Assignment Help Chemistry Assignment Help
Chemistry Assignment Help
 
Solution of nonlinear_equations
Solution of nonlinear_equationsSolution of nonlinear_equations
Solution of nonlinear_equations
 
Linear programming in computational geometry
Linear programming in computational geometryLinear programming in computational geometry
Linear programming in computational geometry
 
Implicit schemes for wave models
Implicit schemes for wave modelsImplicit schemes for wave models
Implicit schemes for wave models
 
signal and system Dirac delta functions (1)
signal and system Dirac delta functions (1)signal and system Dirac delta functions (1)
signal and system Dirac delta functions (1)
 
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8Giáo trình Phân tích và thiết kế giải thuật - CHAP 8
Giáo trình Phân tích và thiết kế giải thuật - CHAP 8
 
SURF 2012 Final Report(1)
SURF 2012 Final Report(1)SURF 2012 Final Report(1)
SURF 2012 Final Report(1)
 
Linear programming in computational geometry
Linear programming in computational geometryLinear programming in computational geometry
Linear programming in computational geometry
 
Project in Calcu
Project in CalcuProject in Calcu
Project in Calcu
 
Design and Analysis of Algorithms Exam Help
Design and Analysis of Algorithms Exam HelpDesign and Analysis of Algorithms Exam Help
Design and Analysis of Algorithms Exam Help
 
Limits and continuity powerpoint
Limits and continuity powerpointLimits and continuity powerpoint
Limits and continuity powerpoint
 

Review

  • 1. MIDTERM REVIEW SESSION FOR CS261 1. What's this about? approximation algorithm • in theory: a rigorous way to deal with NP-hardness. focus on worst-case guarantee. in practice: even a 3/4-approximation is unsatisfactory. cutting plane methods. integer program solver. local search. heuristics. • common: both often use optimal LP value as a benchmark. • there's always a gap between theory and practice. 2. Combinatorial Approach Vertex Cover: unnatural greedy is 2-approx. size of max matching as a lower bound. Steiner Tree: non-metric version reducible to the metric version. then do (1) double the tree (dfs) (2) traverse (3) shortcut (justied by metric ineq.). TSP: general cost + non-repetition is not O(1)-approximable. if you can repeat, then metric is ne. • 2-approx: double the MST + shortcut. • 3/2-approx: MST + min-cost perfect matching on odd-degree nodes (at cost=half of optimal) + eulerian cycle + shortcut. Set Cover: has VC as a special case. L.B.: if we have a subset D of elements, and every set Si contains at most t elements of D, then opt is at least |D|/t. • when the i-th element xi is covered, ci new elements were covered. 1 amortize the cost: cost(xi ) = ci • opt= i cost(xi ). • every set covers at most ci of the n − i + 1 elements. So we have a l.b. n−i+1 on opt: opt ≥ ci (n − i + 1) · cost(xi ) 3. Linear Programming 3.1. LP. primal: max cT x subject to Ax ≤ b, x ≥ 0. (called packing LP if A ≥ 0) dual: min bT y subject to AT y ≥ c, y ≥ 0. (called covering LP if A ≥ 0) (more often primal is minimization problem, then dual is the maximiza- tion problem.) outcome: feasible. infeasible. unbounded. meaning: highest lower-bound on primal optimal. weak duality: primal opt ≥ dual opt strong duality: primal opt = dual opt, except when both infeasible. duality in form: inequality ↔ nonnegativity, equality ↔ unconstrained. complexity: solvable in PTIME. even simplex does well. 1
  • 2. MIDTERM REVIEW SESSION FOR CS261 2 3.2. ILP and LP Relaxations. ILP: linear constraints + integer variables. often NP-hard. (still often fea- sible in practice. using cplex) LP Relaxation: simply replace integrality constraint to like x ≥ 0. What's A Formulation: It's kinda fuzzy. But when you use it, it needs ideally: feasible solution to ILP ↔ feasible solution to problem also work: optimal solution to ILP ↔ optimal solution to problem bottomline: the LP optimal value should give a bound on optimal to the problem.(and hopefully not too far away) Lower Bounds: • the optimal value of LP relaxation fractional optimal • objective value of any feasible dual solution 3.3. Weighted Vertex Cover. LP Rounding: solve LP. pick the vertex i xv ≥ 1/2. (1. ensures it gives a VC, 2. lose by a factor of 2). Primal Dual: dual yu,v . • old alg: if (u,v) uncovered, set xu , xv , yu,v = 1. matching cast as a y(u,v) = 1 if (u,v) in M. dual sol: set • use twice of dual y(u,v) to construct payments pv for costs of vertices, while preserving dual feasibility. pick v if fully-paid for. 3.4. Weighted Set Cover. LP Rounding: we round w.p. x∗ i independently, may not be a cover. but each element with probability at least 1-1/e covered. x: repeat until all elements covered. after log iterations, still exists uncovered elements with exponentially small probability. Primal Dual: eective cost of a set cj = wi newlycovered . same logic as before. in fact dual pj =eective cost. • in fact, the primal is fully-paid for by eective payments, scaled down 1 by H(m) is feasible: H(m) j∈S pj ≤ c(S). 4. Max Flow / Min Cut 4.1. Basic Concepts. input: directed graph with capacities c : E → [0, ∞) and s, t pair. ow: f : E → [0, ∞) s.t. (1) upper-bounded by capacity, and (2) satises conservation constraints. size of a ow: sum of f from source. convention: ow/capacity cut: partition (A, V A) s.t. s ∈ A, t ∈ A. c(A) = u∈A,v∈A,(u,v)∈E c(u, v). / / inequalities: for any s-t cut A: cost(f ) = v∈V :(s,v)∈E f (s, v) = f (A) ≤ a∈A,b∈A,(a,b)∈E f (u, v) ≤ / a∈A,b∈A,(a,b)∈E c(u, v) = c(A). / witness of optimality: a cut A that's saturated. 4.2. Ford-Fulkerson. residual network: capacity = how much more ow you can push. augmenting path: s-t path with positive capacity arcs. Ford-Fulkerson: while an augmenting path can be found, push the maxi- mum amount of ow on the path.
  • 3. MIDTERM REVIEW SESSION FOR CS261 3 Termination: let A be set of nodes to which you can have augmenting path to from s. the cut across A is saturated: f (a, b) = c(a, b). and f (b, a) = 0. Implication: Max-Flow = Min-Cut time complexity: as bad as |C| for bad choices of aug paths. 5. On Writing Solutions • Common Main ingredients: Formulation, Justication, Algorithm, Runtime, Feasibility, Approxi- mation Formulation: make all quanties explicit Justication: you need to show that LP optimal gives a bound on optimal. Algorithm: Explicitly state your algorithm. Runtime: State it. If you need to solve LP exactly, mention that this can be done in PTIME. Feasibility: Show it. Approximation: Be explicit about all important quantities, and the inequal- ities. be explict if you use weak duality. • Miscellaneous Stu taking dual is a mechanical process. don't be too smart. (optional) please don't use matrix notations when you are asked to write LPs. be explicit about whether you round it independently or dependently. be explicit about application of linearity of expectation. use xv , yu,v etc. not ei,j . no reuse of symbols. randomized rounding is a general method. does not refer to any par- ticular rounding algorithm.