Computer scientist Randal Olsen was trapped inside during a snowstorm last month. With no work, and no chance to leave the house, what was he to do? Read a book? Play Where’s Wally? How about vanquish Where’s Wally once and for all (with science)?
A bored mathematician claimed to have already developed a fool-proof strategy for finding the spectacled man in the striped jumper, but Olsen wanted to one-up this and “carve a trail of defeated Waldo searchers in my Wake”.
Olsen’s plan was to examine all of Wally’s hiding spots and to uncover one ultimate path to follow on any page. This is trickier that it might sound; for 64 hiding places the number of potential paths is 2.5 x 1096, greater than the number of atoms in the known universe!
This is where machine learning was utilised. A computer took a given algorithm and continuously tinkered it, only keeping the results whereby the new algorithm yielded a better path than the previous one. Eventually a path was found which was able to cover the most spots where Wally likes to hang out without backtracking.
- SPONSORED -
So you wanna know what this holy grail of Wally-hunting paths is? Well I’m not going to tell you, because as the developer himself said it’s actually way more fun just to look for him the old fashioned way.