CMSC250

Algorithms have properties

• Runtime
• Space
• Feasibility

------------------

• Well defined
• Independent of means of execution
• Should Terminate

Algorithms have properties

• Runtime

Arithmetic complexity: How many operations (of a certain type) occur

Time complexity: How long does the algorithm run?

Arithmetic complexity: How many operations (of a certain type) occur

Time complexity: How long does the algorithm run?

Best case: fastest or least amount of operations

Worst case: longest or most amount of operations

int minimum(int[] arr){
  int min = arr[0];
  for(int i = 1; i < arr.size(); i++)
    if(arr[i] < min) min = arr[i];
  return min;

How many comparison operations?

How many assignmnets?

Best case, worse case?

Average case?

Average case?

Could mean looking at all possible input and taking the average

Could mean looking at practical input and taking the average

Average case?

int minimum(int[] arr){
  int min = arr[0];
  for(int i = 1; i < arr.size(); i++)
    if(arr[i] < min) min = arr[i];
  return min;

Suppose arr.size() will always be 3

At most 3

What about when $P(arr[i] < min) = \frac{1}{i}$

int minimum(int[] arr){
  int min = arr[0];
  for(int i = 1; i < arr.size(); i++)
    if(arr[i] < min) min = arr[i];
  return min;

What if comparison = 2 secs, assignments = 1 sec

Best case, worse case?

Average case?

int minimum(int[][] arr){
  int min = arr[0][0];
  for(int i = 0; i < arr.size(); i++)
    for(int j = 0; j < arr.size(); j++)
      if(arr[i][j] < min) min = arr[i][j];
  return min;

How many comparisons?

How many assignments best case?

How many assignments worst case?

int minimum(int[][] arr){
  int min = arr[0][0];
  for(int i = 0; i < arr.size(); i++)
    for(int j = 0; j < arr.size(); j++)
      if(arr[i][j] < min) min = arr[i][j];
  return min;

Suppose that $min = A[i,j]$ takes $i + 2^j$ seconds

How long?

int f(int n){
  if(n == 0) return 1;
  return factorial(n)+ f(n-1); 

How many additions?

How many multiplications?

Note: different runtimes same algorithm

Algorithm can be broken up

int f(int n){
  if(n == 0) return 1;
  return factorial(n)+ f(n-1); 

Addition takes 2 seconds, multiplication 6 seconds

Suppose input is a equally likely random number between 1 and 100 inclusive

Average runtime? (aka expected value?)