Hi,
Bit of a long shot but im wondering if anyone knows how to do this.
x could have all possible integer values from 0<x<63 (inclusive) with a uniform distribution.
The entropy of x, H(x) = 6 bits.
Now given this code:
I need to work out the 'mutual information' between y and x. i.e. How much information does y reveal about x.
It's defined as I(Y;X) = H(Y) - H(Y|X)
I don't really understand how to apply conditional entropy to this problem though.
Anyone have any ideas?
Bit of a long shot but im wondering if anyone knows how to do this.
x could have all possible integer values from 0<x<63 (inclusive) with a uniform distribution.
The entropy of x, H(x) = 6 bits.
Now given this code:
Code:
int y;
int r = rand(0, 1); // random integer 0 or 1
y = r * 64 + x;I need to work out the 'mutual information' between y and x. i.e. How much information does y reveal about x.
It's defined as I(Y;X) = H(Y) - H(Y|X)
I don't really understand how to apply conditional entropy to this problem though.
Anyone have any ideas?