Complexity course outline Spring 1997

Lectures given by Prof. Ron Shamir
Recitations given by Zvi Ostfeld.

Date Type Topics reference handouts
Feb. 17 Lecture 1 Introduction, Algorithms, growth rate of fuctions. Reachability; Max Flow; Matching; The Travelling Salesman Problem. P chapter 1, GJ 1.3 whitepaper and syllabus
Feb. 19 Targil* Survey about orders of magnitude and Turing machines. P chapter 2. Home assignment #1.
Feb. 20 Lecture 2 Turing Machines: Model, time complexity, k-tape machine, simulation by 1-tape machine; linear speedup theorem; space complexity P 2.1-2.5 my notes
Feb 24 Lecture 3 Random Access Machine: definition and examples. Polynomial equivalence of RAM and TM. Nondeterministic Machines (incomplete) P 2.6 my notes
Feb. 26 Targil A RAM program to solve the reachability problem. Few examples of Turing machine programs. P chapter 2. P pages 41-42. Home assignment #2.
Feb 27 Lecture 4 nondeterministic machines, proper complexity functions, complexity classes 2.7, 7.1 -
March 3 Lecture 5 complementary classes, universal Turing machines, Time and Space Hierarchy theorems, Gap theorem (w/o proof), relations between deterministic and nondeterministic complexity classes (incomplete); 7.1 7.2 -
Mar. 5 Targil TM program for addition and subtraction. Survey about flow problems. Brief survey of the following classes: P, NP and CO-NP. - Home assignment #3.
March 6 Lecture 6 relations between deterministic and nondeterministic complexity classes, Savitch's theorem; the configuration graph, the reachability method, Immerman's theorem. 7.3 -
March 10 Lecture 7 Reductions; Hamlitonian Path < SAT; Circuit SAT < SAT; reduction by generalization 8.1 -
Mar. 12 Targil SAT < 3SAT - Home assignment #4.
March 13 Lecture 8 Transitivity of (logspace) reductions; Completeness; Circuit Value is P-complete (two proofs) 8.1 8.2, JaJa 10.5 Outline of JaJa's reduction
March 17 Lecture 9 Cook's Theorem (2 proofs) variation of SAT: 3SAT, 3 occurences per variable. SAT < Independet Set; SAT < Hypergraph 2-colorability 8.2 9.1 Even 9.3 Outline of Cook's theorem a-la Even
Mar. 19 Targil 3SAT < 3COLORING; A language decided by a k-string NDTM can be decided a 2-string NDTM without affecting the time complexity. - Home assignment #5.
March 20 Lecture 10 Hypergraph 2-colorability: restrictions; Set Splitting; Not-All-Equal SAT; 3SAT < Directed Hamiltonian Cycle DHC reduction from Hopcroft-Ullman simplified figure for the DHC reduction
March 24 Lecutre 11 3SAT < 3DM , X3C GJ 3.1.2 3DM reduction summary and sketch
Mar. 26 Targil Reductions; Reachability < Boolean Circuit ; CVP < Liner Inuequality ; 3SAT < MAX2SAT 8.2 ; JaJa p. 552 Home assignment #6.
March 27 Lecture 12 Characterization of NP by poly. balanced and poly. decidable relations NAE-3SAT < Max Cut; Max Bisection, Bisection width. X3C < Subset Sum; Subset Sum < Partition. P 9.1; 9.2; GJ 3.1.5 simplified -
Mar 31 Lecture 13 Dynamic programming algorithm for Subset Sum; Number problems; Pseudopolynomial algorithms; strong NP-completeness; Pseudopolynomial reductions; GJ 4.2 -
Apr. 2 Targil Reductions: Few variants of UHP and UHC - Home assignment #7.
Apr. 3 Lecture 14 Bin Packing is Strongly NP-complete (redcution from 3DM); 3-Partition (w/o proof); Interval Sequencing: Pseudo polynomial reduction from 3-partition P 9.4, GJ 4.2.2 -
Apr. 7 Lecutre 15 Subgraph Isomorphism is NPC; Subforest isomorphism: pseudo-polynomial reduction from 3-partition.
Dynamic Programming: order of matrix multiplication; TSP; Minimum edit distance between sequences (incomplete)
GJ 4.2.2,
AHU 2.8, BB p. 159; Gusfield's text
Minimum edit distance example from Gusfield's text.
Apr. 9 Targil Few variants of UHP and UHC (cont'); Restricted version of Clique (which is NPC); Restricted version of VC (which is polynomial) - Home assignment #8.
Apr. 10 Lecture 16 Minimum edit distance between sequences (completed); coNP, the intersection of NP and coNP, Primality and Pratt's theorem; Function problems (incomplete) P 10 -
Apr. 14 Lecture 17 Function problems, FP, FNP, reductions between function problems;
Oracle Turing Machines, Turing reductions, NP-hardness;
Approximation: 1/2-approximation for Bin Packing; First Fit, First Fit Decreasing, Best Fit (w/o proofs);
P 10,
P 14.3, GJ 5.1;
P 13.1, GJ 6.1
-
Apr. 30 Targil Difference between Log Space Reduduction, Karp Reduction and Cook (Turing) Reduction - Home assignment #9.
May. 1 Lecture 18 1/2-approximation for vertex cover. TSP with triangle inequality: 1/3-approximation algorithm. FPTAS for Knapsack. P 13.1, GJ 6.1 -
May. 5 Lecture 19 Approximation ratios; Polynomial and fully polynomial time approximation schemes, Absolute approximation for coloring planar graphs. Hardness of approximation: Nonexistence of Absolute approximation for Knapsack, fixed ration for TSP. L-reductions, properties, MAXSNP; GJ 6.1, P. 13.1, 13.2 -
May. 7 Targil Polinomial algorithm for coloring a 3-Colorabale graph with O(sqrt |V|) colors; A polinomial 1.5-epsilon approximation algorithm for VC implies a polinomial algorithm for coloring a 3-Colorabale graph with O(log |V|) colors; A 1/2-approximation algorithm for Knapsack; Wi83 Home assignment #10.
May 8 Lecture 20 MAXSNP, MAXSNP-completeness; the new (PCP) results on hardness of approximations (w/o proofs). Minimum Equivalent Expression; The Polynomial Heirarchy; Complete problems for levels of the PH. Backtracking algorithms: n queens, 3 cloroing. GJ 6.1, P. 13.2 -
May. 15 Lecture 21 Backtracking algorithms: n queens, 3 cloroing, Hamiltonian Cycle. Examples of branch and bound heuristics: TSP, Knapsack.
PSPACE; The Quantified Boolean Formula (QBF) problem;
Sedgewick, BB p. 200
P 19.1
Backtracking example for Hamiltonian Cycle (Sedgewick) Home assignment #11.
May 19 Lecture 22 QBF is PSPACE complete; Alternating Turing Machines, AP, AL. AL=P; AP=PSPACE. P 19.1; P. 16.2, 19.1, 19.4 -
May 22 Lecture 23 More PSPACE-complete problems: Geography; In-Place Acceptance.
Counting problems: The class #P, parsimonious reductions; #SAT is #P-complete.
P. 19.4,
P. 18.1
-
May 26 Lecture 24 Perfect matchings and the permanent. Valiant's Theorem (w/o proof). The class PP; Toda's theorem (w/o proof).
Parallel Computations. The PRAM model and its variants. Finding the minimum of n elements in constant parallel time.
P. 18.1
-
-
May. 28 Targil An existence of a polynomial algorithm that achieves a constant ratio when approximating the maximum clique implies the existence of a polynomial algorithm that achieves 1 + epsilon ratio when approximating the maximum clique for every constant epsilon > 0. Travelling salesman optimization problem is NP-easy. - Home assignment #12.
May 29 Lecture 25 Parallel algorithms for matrix multiplications, Reachability. Parallel computations models: Uniform Boolean Circuits, PRAM. Simulation Theorems (w/o proofs). P. 15.1 15.2 -
Jun. 4 Targil The polynimial hierarchy. Turing reductions and Karp reductions. - Home assignment #13.
June 5 Lecture 26 NC; Logarithmic space; In pursuit of the P?NP question (review) GJ 7.6 P 14 -

*Targil=recitation (in Hebrew)

< = reduces to (using logarithmic space deterministic Turing Machine)

P= C. Papadimitriou Computational Complexity (Addison Wesley, 1994)
GJ= Garey and Johnson, Computers and Intractability (Freeman, 1979)
AHU= Aho, Hopcroft, and Ullman. The Design and Analysis of Computer Algorithms . (Addison Wesley, Reading, Mass., 1974).
BB= Brassard and Bratley. Algorithmics: Theory and Practice . (Prentice Hall, Englewood Cliffs, N. J., 1988).
CLR= Cormen, Leiserson and Rivest, Introduction to Algorithms (MIT press, 1990)
JaJa = J. JaJa. An Introduction to Parallel Algorithms (Addison-Wesley, Reading, Mass., 1992).
Wi83= A. Wigderzon, Improving the performance guarantee for approximate graph coloring, Journal of the ACM,30(4);729-735,1983


Back to Ron's home page