Difference between revisions of "Asymptotic Analysis"

From Suhrid.net Wiki
Jump to navigationJump to search
(Created page with "* Running time of an algorithm is a function of the size of it's input. * How fast does an algorithm grow with input size ? This is the rate of growth of the running time. *...")
 
 
Line 2: Line 2:
 
* How fast does an algorithm grow with input size ? This is the rate of growth of the running time.  
 
* How fast does an algorithm grow with input size ? This is the rate of growth of the running time.  
 
* Three kinds of asymptotic notation :
 
* Three kinds of asymptotic notation :
* Big Theta : Provides an asymptotically tight bound on the running time of an algo. If f(n) is BigTheta(g(n)) this means that f(n) grows asymptotically at the same rate as g(n)
+
* Big Theta : Provides an asymptotically tight bound on the running time of an algo. If f(n) is BigTheta(g(n)) this means that f(n) grows (asymptotically) at the same rate as g(n)
* Big Omega : Provides an asymptotically lower bound on the running time of an algo. If f(n) is BigOmega(g(n)) this means that f(n) grows asymptotically no slower than g(n)
+
* Big Omega : Provides an asymptotically lower bound on the running time of an algo. If f(n) is BigOmega(g(n)) this means that f(n) grows (asymptotically) no slower than g(n)
* Big O : Provides an asymptotically upper bound on the running time of an algo. If f(n) is O(g(n)) this means that f(n) grows asymptotically no faster than g(n)
+
* Big O : Provides an asymptotically upper bound on the running time of an algo. If f(n) is O(g(n)) this means that f(n) grows (asymptotically) no faster than g(n)
 
* e.g. consider binary search, the worst-case running time is log(n).  
 
* e.g. consider binary search, the worst-case running time is log(n).  
 
* So we can say Running time of binary search i.e. f(n) - the actual algorithm logic - grows no faster than log(n) i.e. g(n). So we say BinarySearch is O(log(n)).
 
* So we can say Running time of binary search i.e. f(n) - the actual algorithm logic - grows no faster than log(n) i.e. g(n). So we say BinarySearch is O(log(n)).

Latest revision as of 19:45, 4 May 2015

  • Running time of an algorithm is a function of the size of it's input.
  • How fast does an algorithm grow with input size ? This is the rate of growth of the running time.
  • Three kinds of asymptotic notation :
  • Big Theta : Provides an asymptotically tight bound on the running time of an algo. If f(n) is BigTheta(g(n)) this means that f(n) grows (asymptotically) at the same rate as g(n)
  • Big Omega : Provides an asymptotically lower bound on the running time of an algo. If f(n) is BigOmega(g(n)) this means that f(n) grows (asymptotically) no slower than g(n)
  • Big O : Provides an asymptotically upper bound on the running time of an algo. If f(n) is O(g(n)) this means that f(n) grows (asymptotically) no faster than g(n)
  • e.g. consider binary search, the worst-case running time is log(n).
  • So we can say Running time of binary search i.e. f(n) - the actual algorithm logic - grows no faster than log(n) i.e. g(n). So we say BinarySearch is O(log(n)).