2 min read
logarithm

Related:


[!quote] Math is Fun How many of one number multiply together to make another number ?

Example from Math is Fun1

How many 2’s multiply together to make 8 ?

2 x 2 x 2 = 8

so we need three 2’s to get a product of 8

we’d write it as the following

$\log_{2}(8)=3$

key terms

  • base: number we multiply (example: 2)
  • logarithm: how often the base needs to be multiplied (example: 3)
  • the desired number (example: 8)

Common logarithm (log) are base 10. If they’re base 10, they don’t need to precise the base number

natural logarithms have a base e, also known as [[Euler’s number|Euler’s number]]

computer science - algorithms and logs

[[../../../dichot algo example - ED 4|dichot]] (Binary Search) Time Complexity: $O(log⁡ n)$ Quick Explanation: This algorithm repeatedly divides the search interval in half, leading to logarithmic time complexity. Detailed explanation:

  • the code has one loop
  • each iteration of the loop results in the vector length being cut in half
    • using logarithm, this can be expressed as $\log_{2}(n)$ = how many times I can divide by 2 before getting to 1
    • where $n$ are the number of elements in a vector pool
  • the reason why it’s defined as $log(n)$ and not $\log_{2}(n)$ is because in computer science log usually means $\log_{2}$ [^2] due to ignoring constant factors
    • so if the algorithm divides the problem by 2, 3, or 10 at each step, the growth characteristic is still logarithmic
  • so time complexity of this algorithm in $O log(n)$

References

Footnotes

  1. Introduction to Logarithms [^2] Time complexity binary search