What complexity have a log(sum) shape?

Timon Gehr via Digitalmars-d digitalmars-d at puremagic.com
Tue Dec 8 16:05:15 PST 2015


On 12/08/2015 11:31 PM, Andrei Alexandrescu wrote:
> On 12/08/2015 04:55 PM, Timon Gehr wrote:
>> The context is asymptotic runtime bounds. The special cases for small
>> values of continuous logarithms can just be defined away. They have no
>> relevance for asymptotic runtime analysis (even though defining big-O
>> adequately for multiple variables is more tricky than for a single
>> variable).
>
>  From the cycle "Andrei's esprits d'escalier", the inequality can be
> derived from the somewhat notorious log sum inequality if you make all
> bi equal:
> http://ocw.usu.edu/Electrical_and_Computer_Engineering/Information_Theory/lecture3.pdf
>
>
> Andrei

Which inequality?


More information about the Digitalmars-d mailing list