1 / 67

CHAPTER 2 Analysis of Algorithms

CHAPTER 2 Analysis of Algorithms. Input. Algorithm. Output. An algorithm is a step-by-step procedure for solving a problem in a finite amount of time. Algorithm Analysis: Do NOT worry !!!. Math you need to Review. Series summations Logarithms and Exponents Proof techniques

ahanu
Télécharger la présentation

CHAPTER 2 Analysis of Algorithms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CHAPTER 2Analysis of Algorithms Input Algorithm Output An algorithm is a step-by-step procedure for solving a problem in a finite amount of time.

  2. Algorithm Analysis: Do NOT worry !!! Analysis of Algorithms

  3. Math you need to Review • Series summations • Logarithms and Exponents • Proof techniques • Basic probability • properties of logarithms: logb(xy) = logbx + logby logb (x/y) = logbx - logby logbxa = alogbx logba = logxa/logxb • properties of exponentials: a(b+c) = aba c abc = (ab)c ab /ac = a(b-c) b = a logab bc = a c*logab Analysis of Algorithms

  4. Sum series-examples Analysis of Algorithms

  5. Analysis of Algorithms

  6. Analysis of Algorithms

  7. Recurrence Relations Analysis of Algorithms

  8. Analysis of Algorithms

  9. Analysis of Algorithms

  10. Analysis of Algorithms

  11. Analysis of Algorithms

  12. Analysis of Algorithms

  13. Assumptions for the computational model • Basic computer with sequentially executed instructions • Infinite memory • Has standard operations; addition, multiplication, comparison in 1 time unit unless stated otherwise • Has fixed-size (32bits) integers such that no-fancy operations are allowed!!! eg.matrix inversion, • Proof techniques • Basic probability Analysis of Algorithms

  14. Running Time • Most algorithms transform input objects into output objects. • The running time of an algorithm typically grows with the input size. • Average case time is often difficult to determine. • We focus on the worst case running time. • Easier to analyze • Crucial to applications such as games, finance, image,robotics, AI, etc. Analysis of Algorithms

  15. Experimental Studies • Write a program implementing the algorithm • Run the program with inputs of varying size and composition • Assume a subprogram is written to get an accurate measure of the actual running time • Plot the results Analysis of Algorithms

  16. Limitations of Experiments • It is necessary to implement the algorithm, which may be difficult • Results may not be indicative of the running time on other inputs not included in the experiment. • In order to compare two algorithms, the same hardware and software environments must be used Analysis of Algorithms

  17. Theoretical Analysis • Uses a high-level description of the algorithm instead of an implementation • Characterizes running time as a function of the input size, n. • Takes into account all possible inputs • Allows us to evaluate the speed of an algorithm independent of the hardware/software environment Analysis of Algorithms

  18. Example: find max element of an array AlgorithmarrayMax(A, n) Inputarray A of n integers Outputmaximum element of A currentMaxA[0] fori1ton  1do ifA[i]  currentMaxthen currentMaxA[i] returncurrentMax Pseudocode • High-level description of an algorithm • More structured than English prose • Less detailed than a program • Preferred notation for describing algorithms • Hides program design issues Analysis of Algorithms

  19. Control flow if…then… [else…] while…do… repeat…until… for…do… Indentation replaces braces Method declaration Algorithm method (arg [, arg…]) Input… Output… Method call var.method (arg [, arg…]) Return value returnexpression Expressions Assignment(like  in Java) Equality testing(like  in Java) n2 Superscripts and other mathematical formatting allowed Pseudocode Details Analysis of Algorithms

  20. Important Functions • Seven functions that often appear in algorithm analysis: • Constant  1 • Logarithmic  log n • Linear  n • N-Log-N  n log n • Quadratic  n2 • Cubic  n3 • Exponential  2n • In a log-log chart, the slope of the line corresponds to the growth rate of the function Analysis of Algorithms

  21. Basic computations performed by an algorithm Identifiable in pseudocode Largely independent from the programming language Exact definition not important (we will see why later) Assumed to take a constant amount of time in the RAM model Examples: Evaluating an expression Assigning a value to a variable Indexing into an array Calling a method Returning from a method Primitive Operations Analysis of Algorithms

  22. Detailed Model 4 counting operations-1 The time require for the following operations are all constants. Analysis of Algorithms

  23. Detailed Model 4 counting operations-2 The time require for the following operations are all constants. Analysis of Algorithms

  24. Detailed Model 4 counting operations-3 Analysis of Algorithms

  25. Detailed Model... Example1:Sum ... public static int sum(int n) { int result = 0; for (int i = 1; i <= n; ++i){ result += i; } return result; } ... Analysis of Algorithms

  26. Detailed Model... Example2 (*) y = a [i] 3fetch + [.] + store • operations • fetch a (the base address of the array) • fetch i (the index into the array) • address calculation • fetch array element a[i] • store the result Analysis of Algorithms

  27. publicclass Horner { publicstaticvoid main(String[] args) { Horner h = new Horner(); int[] a = { 1, 3, 5 }; System.out.println("a(1)=" + h.horner(a, a.length - 1, 1)); System.out.println("a(2)=" + h.horner(a, a.length - 1, 2)); } int horner(int[] a, int n, int x) { int result = a[n]; for (int i = n - 1; i >= 0; --i) { result = result * x + a[i]; /**/System.out.println("i=" + i + " result" + result); } return result; } } i=1 result8 i=0 result9 a(1)=9 i=1 result13 i=0 result27 a(2)=27 output Detailed Model...Example3: Horner (*) Analysis of Algorithms

  28. publicclass FindMaximum { publicstaticvoid main(String[] args) { FindMaximum h = new FindMaximum(); int[] a = { 1, 3, 5 }; System.out.println("max=" + h.findMaximum(a)); } int findMaximum(int[] a) { int result = a[0]; for (int i = 0; i < a.length; ++i) { if (result < a[i]) { result = a[i]; } System.out.println("i=" + i + " result=" + result); } return result; } } i=0 result=1 i=1 result=3 i=2 result=5 max=5 output Detailed Model...Example-4 (*) 3fetch + [.]+ store fetch + store (2fetch + <) n (2fetch + ++ store) (n-1) (4fetch + [.]+ <) (n-1) (3fetch + [.]+ store) ? fetch + store Analysis of Algorithms

  29. Simplified Model … More Simplification • All timing parameters are expressed in units of clock cycles. In effect, T=1. • The proportionality constant, k, for all timing parameters is assumed to be the same: k=1. Analysis of Algorithms

  30. publicclass FindMaximum { publicstaticvoid main(String[] args) { FindMaximum h = new FindMaximum(); int[] a = { 1, 3, 5 }; System.out.println("max=" + h.findMaximum(a)); } int findMaximum(int[] a) { int result = a[0]; for (int i = 0; i < a.length; ++i) { if (result < a[i]) { result = a[i]; } System.out.println("i=" + i + " result=" + result); } return result; } } i=0 result=1 i=1 result=3 i=2 result=5 max=5 out Simplified Model …Example_FindMax 1 2 3 4 5 6 7 8 9 10 detailed 2 3fetch + [.]+ store 3a fetch + store 3b (2fetch + <) n 3c (2fetch + ++ store) (n-1) 4 (4fetch + [.]+ <) (n-1) 6 (3fetch + [.]+ store) ? 9 fetch + store simple 2 5 3a 2 3b (3)n 3c (4)(n-1) 4 (6)(n-1) 6 (5)? 9 2 Analysis of Algorithms

  31. publicclass GeometrikSeriesSum { publicstaticvoid main(String[] args) { System.out.println("1, 4: " + geometricSeriesSum(1, 4)); System.out.println("2, 4: " + geometricSeriesSum(2, 4)); } publicstaticint geometricSeriesSum(int x, int n) { int sum = 0; for (int i = 0; i <= n; ++i) { int prod = 1; for (int j = 0; j < i; ++j) { prod *= x; } sum += prod; } return sum; } } 1, 4: 5 2, 4: 31 output Simplified Model > Algorithm1…Geometric Series Sum 1 2 3 4 5 6 7 8 9 10 11 12 simple 1 2 2 3a 2 3b 3(n+2) 3c 4(n+1) 4 2(n+1) 5a 2(n+1) 5b 3 (i+1) 5c 4 i 6 4 i 7 8 4(n+1) 9 10 2 11 12 Total 11/2 n2 + 47/2 n +27 Analysis of Algorithms

  32. publicclass GeometrikSeriesSumHorner { publicstaticvoid main(String[] args) { System.out.println("1, 4: " + geometricSeriesSum(1, 4)); System.out.println("2, 4: " + geometricSeriesSum(2, 4)); } publicstaticint geometricSeriesSum(int x, int n) { int sum = 0; for (int i = 0; i <= n; ++i) { sum = sum * x + 1; } return sum; } } 1, 4: 5 2, 4: 31 output Simplified Model > Algorithm_SumHornerGeometric Series Sum … 1 2 3 4 5 6 7 • 2 2 • 3a 2 • 3b 3(n+2) • 3c 4(n+1) • 4 6(n+1) • 6 2 • Total 13 n + 22 Analysis of Algorithms

  33. publicclass GeometrikSeriesSumPower { publicstaticvoid main(String[] args) { System.out.println("1, 4: " + powerA(1, 4)); System.out.println("1, 4: " + powerB(1, 4)); System.out.println("2, 4: " + powerA(2, 4)); System.out.println("2, 4: " + powerB(2, 4)); } ... publicstaticint powerA(int x, int n) { int result = 1; for (int i = 1; i <= n; ++i) { result *= x; } return result; } publicstaticint powerB(int x, int n) { if (n == 0) { return 1; } elseif (n % 2 == 0) { // n is even return powerB(x * x, n / 2); } else { // n is odd return x * powerB(x * x, n / 2); } } } 1, 4: 1 1, 4: 1 2, 4: 16 2, 4: 16 output Simplified Model > Algorithm_SumPower Geometric Series Sum … 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 powerB  1 n=0 xn =  (x2)n/2 0<n, n is even  x(x2)n/2 0<n, n is odd powerB 0<n 0<n n=0 n even n odd 10 3 3 3 11 2 - - 12 - 5 5 13 - 10+T(n/2) - 15 - - 12+T(n/2) Total 5 18+T(n/2) 20+T(n/2) powerB  5 n=0 xn =  18+T(n/2)0<n, n is even  20+T(n/2)0<n, n is odd Analysis of Algorithms

  34. Let n = 2k for some k>0. Since n is even, n/2 = n/2 = 2k-1. For n = 2k, T(2k) = 18 + T(2k-1). Using repeated substitution T(2k) = 18 + T(2k-1) = 18 + 18 + T(2k-2) = 18 + 18 + 18 + T(2k-3) … = 18j + T(2k-j) Substitution stops when k = j T(2k) = 18k + T(1) = 18k + 20 + T(0) = 18k + 20 + 5 = 18k + 25. n = 2k then k = log2 n T(2k) = 18 log2 n + 25 Simplified Model > Calculation of PowerB …Geometric Series Sum … (*) n is even powerB xn =  18+T(n/2)0<n Analysis of Algorithms

  35. Simplified Model > Calculation of PowerB Geometric Series Sum … (*) Suppose n = 2k–1 for some k>0. Since n is odd, n/2 = (2k–1)/2 = (2k–2)/2= 2k-1 For n = 2k-1, T(2k-1) = 20 + T(2k-1-1), k>1. Using repeated substitution T(2k-1) = 20 + T(2k-1-1) = 20 + 20 + T(2k-2-1) = 20 + 20 + 20 + T(2k-3-1) … = 20j + T(2k-j-1) Substitution stops when k = j T(2k-1) = 20k + T(20-1) = 20k + T(0) = 20k + 5. n = 2k-1then k = log2 (n+1) T(n) = 20 log2 (n+1) + 5 n is odd Therefore, powerB  5 n=0 xn =  18+T(n/2)0<n, n is even  20+T(n/2)0<n, n is odd Average: 19(log2(n+1) + 1) + 18 Analysis of Algorithms

  36. publicclass GeometrikSeriesSumPower { publicstaticvoid main(String[] args) { System.out.println(“s 2, 4: " + geometrikSeriesSumPower (2, 4)); } ... publicstaticint geometrikSeriesSumPower (int x, int n) { return powerB(x, n + 1) – 1 / (x - 1); } publicstaticint powerB(int x, int n) { if (n == 0) { return 1; } elseif (n % 2 == 0) { // n is even return powerB(x * x, n / 2); } else { // n is odd return x * powerB(x * x, n / 2); } } } s 2, 4: 31 output Simplified Model > Algorithm_SumPower Geometric Series Sum … Analysis of Algorithms

  37. Comparison of 3 Algorithms algorithm T(n) Sum 11/2 n2 + 47/2 n + 27 Horner 13n + 22 Power 19(log2(n+1) + 1) + 18 sum Horner Power Analysis of Algorithms

  38. By inspecting the pseudocode, we can determine the maximum number of primitive operations executed by an algorithm, as a function of the input size AlgorithmarrayMax(A, n) # operations currentMaxA[0] 2 fori1ton 1do 2n ifA[i]  currentMaxthen 2(n 1) currentMaxA[i] 2(n 1) { increment counter i } 2(n 1) returncurrentMax 1 Total 8n 2 Counting Primitive Operations for Pseudocodes Analysis of Algorithms

  39. Algorithm arrayMax executes 8n 2 primitive operations in the worst case. Define: a = Time taken by the fastest primitive operation b = Time taken by the slowest primitive operation Let T(n) be worst-case time of arrayMax.Thena (8n 2) T(n)b(8n 2) Hence, the running time T(n) is bounded by two linear functions Estimating Running Time Analysis of Algorithms

  40. Asymptotic Algorithm Analysis • The asymptotic analysis of an algorithm determines the running time in big-Oh notation • To perform the asymptotic analysis • We find the worst-case number of primitive operations executed as a function of the input size • We express this function with big-Oh notation • Example: • We determine that algorithm arrayMax executes at most 8n 2 primitive operations • We say that algorithm arrayMax “runs in O(n) time” • Since constant factors and lower-order terms are eventually dropped anyhow, we can disregard them when counting primitive operations Analysis of Algorithms

  41. Growth Rate of Running Time • Changing the hardware/ software environment • Affects T(n) by a constant factor, but • Does not alter the growth rate of T(n) • The linear growth rate of the running time T(n) is an intrinsic property of algorithm arrayMax Analysis of Algorithms

  42. Constant Factors • The growth rate is not affected by • constant factors or • lower-order terms • Examples • 102n+105is a linear function • 105n2+ 108nis a quadratic function Analysis of Algorithms

  43. Big-Oh Notation • Given functions f(n) and g(n), we say that f(n) is O(g(n))if there are positive constantsc and n0 such that f(n)cg(n) for n n0 • Example: 2n+10 is O(n) • 2n+10cn • (c 2) n  10 • n  10/(c 2) • Pick c = 3 and n0 = 10 Analysis of Algorithms

  44. Big-Oh Example • Example: the function n2is not O(n) • n2cn • n c • The above inequality cannot be satisfied since c must be a constant Analysis of Algorithms

  45. More Big-Oh Examples • 7n-2 7n-2 is O(n) need c > 0 and n0 1 such that 7n-2  c•n for n  n0 this is true for c = 7 and n0 = 1 • 3n3 + 20n2 + 5 3n3 + 20n2 + 5 is O(n3) need c > 0 and n0 1 such that 3n3 + 20n2 + 5  c•n3 for n  n0 this is true for c = 4 and n0 = 21 • 3 log n + 5 3 log n + 5 is O(log n) need c > 0 and n0 1 such that 3 log n + 5  c•log n for n  n0 this is true for c = 8 and n0 = 2 Analysis of Algorithms

  46. Big-Oh and Growth Rate • The big-Oh notation gives an upper bound on the growth rate of a function • The statement “f(n) is O(g(n))” means that the growth rate of f(n) is no more than the growth rate of g(n) • We can use the big-Oh notation to rank functions according to their growth rate Analysis of Algorithms

  47. More Example Analysis of Algorithms

  48. Example Analysis of Algorithms

  49. Big-Oh Rules Theorem Consider polynomial where am>0.Then f(n) = O(nm). i.e., • Drop lower-order terms • Drop constant factors • Use the smallest possible class of functions • Say “2n is O(n)”instead of “2n is O(n2)” • Use the simplest expression of the class • Say “3n+5 is O(n)”instead of “3n+5 is O(3n)” Analysis of Algorithms

  50. Big-Oh Rules-2 • logk n = O(n) for any constant kZ+ • f(n) = O (f(n)) • c O(f(n)) = O (f(n)) • O(f(n)) + O(f(n)) = O(f(n)) • O(O(f(n))) = O(f(n)) • O(f(n)) O(g(n)) = O(f(n) g(n)) Analysis of Algorithms

More Related