It’s no big deal, I’m just going to use this algorithm to find you. (Photo: Candlewick Press)
There’s now no longer a need to spend hours poring over “Where’s Waldo?” illustrations to find that slippery cartoon hipster. There’s an algorithm for that.
According to a report in the Guardian, a doctoral student at Michigan State University’s High-Performance Computing Center set out to defeat the popular puzzle book series, created in 1991, once and for all.
The student, Randy Olson, explained that, after being snowed in on the first weekend of February, he challenged himself to the task:
"While searching for something to catch my fancy, I ran across an old Slate article claiming that they found a foolproof strategy for finding Waldo in the classic “Where’s Waldo?” book series,” he wrote on his blog. “Now, I’m no Waldo-spotting expert, but even I could tell that the strategy they proposed there is far from perfect.
"That’s when I decided what my weekend project would be: I was going to pull out every machine learning trick in my tool box to compute the optimal search strategy for finding Waldo. I was going to crush Slate’s supposed foolproof strategy and carve a trail of defeated Waldo-searchers in my wake."
Though Olson’s self-described approach was… startlingly aggressive, his passion led him to construct a rather brilliant strategy for pinning down his red-capped arch nemesis.
He first started by scraping all 68 of Waldo’s coordinates from the first seven editions of “Where’s Waldo?” via a chart provided in the aforementioned article in Slate. He then graphed them on a plane, like so:
Images via randalolson.com
He then applied something called a kernal density estimation — a statistical process that measures probability — to the data. From this, he learned some basic patterns about Waldo’s whereabouts: Waldo is rarely located in the top left corner or in the edges of an illustration. And he is never at the very bottom of the right page.
Olson then applied something called a genetic algorithm, a program that recreates the process of natural selection, to the 68 data points. The program ran through all the exhaustive possibilities of Waldo’s placement until it created an outline for the path of your eye on the page.
Olson assigned a color to each path, based on whether it’s the beginning, middle, or end of where your eye should start. Finally he came up with this model:
So, if you want to find that hippie hitchhiker, you should start at the bottom left corner of the page, then follow this haphazard path across, down to the bottom right side of the illustration. Think of it as a map of America, with a starting point in California, a jaunt up through the Midwest and into New York, and then back down to Georgia.
By the time you’ve made it down South, you’ll have found Waldo, and lost all the wandering imagination you had as a child. Thanks, computers!