Parable of the Hill Climber

Once upon a time, there was a hill climber.

max0001-1

The hill climber wanted to climb the tallest hill. However, the land was foggy and they couldn't see far. So, they climbed in whatever nearby direction seemed uphill.

max0002

They climbed & climbed, and finally reached a peak. “At last,” the hill climber said, “I've climbed the tallest hill!”

max0003

What our protagonist has done is known as the "Hill Climbing" algorithm. Given a point on a landscape of possibilities – where higher points represent better states – the algorithm just looks at nearby points (because evaluating all points would be too costly), moves to the "highest" nearby point, and repeats. Eventually, the algorithm will reach a peak.

Just one problem.

max0004

That peak may not be the peak.

Our hero is now stuck in what's called a "local maximum".

max0005

The hill climber looked to the left. Doesn't seem uphill.

max0006

The hill climber looked to the right. Doesn't seem uphill.

max0007

The hill climber felt trapped. “Is this all there is?” they thought. “Is life literally all downhill from here?”

Little did they know, there are better algorithms. Algorithms like “Stochastic Hill Climbing” and “Simulated Annealing” solve the local-maximum problem by adding randomness. This causes the algorithm to take worse steps in the short-run, so it can locate better states in the long-run.

Sometimes, you have to leave what's been working for you, and take a step down into the fog...

max0008

...to climb to a better place.

max0009


banner

In November, I went on a sabbatical/vacation, to add randomness to my algorithm. I took up sewing, made plushies, studied (very, very intro-level) quantum mechanics, practiced my French in Montréal, hiked through snowy woods, read papers, ate bagels.

I also went to ThinkerCon, where I met hundreds of other internet edu-thing creators! I was on a panel hosted by Robert Krulwich (RadioLab) on exactly the topic of this post: escaping your local maximum.

Vi Hart – awesome human & co-creator of Parable of the Polygons – was on the panel too! Vi's a maestro of escaping local max: first a musician, then YouTuber, then VR researcher, and now getting into AI. It was in talking with Vi I realized:

  1. It's okay to go do something totally different, to avoid local max.

  2. I've hit local max.

Here's my hill: I explain ideas with single-player, mathematical simulations on a flat 2D screen. I am the king/queen of this hill. But what about ideas that can't be explained through mathematical simulations? What about other forms of play, like role-play, multiplayer, play in Virtual/Augmented/Actual Reality? What about purposes other than "explaining", like collaboration, invention, citizen science, storytelling, art?

Point is: I'm ready to step into the fog.

Over the next 6 months, I'll try one new thing a month. (new for me, anyway) In December it'll be a science experiment, where I'll collect & analyze data using an interactive. After that, not sure yet. All I know is that I want to explore new directions. These new directions might seem downhill at first, and honestly most of them might not pan out.

But trying out new directions was how I got here! I went from comics, to animation, to games, to web tools, to now "explorable explanations". I won't leave explorables just yet, but I think there's still so much more good that can be done through the medium of play.

I don't know what's ahead...

...but I'm truly, genuinely grateful –

max0010

– that I get to explore it with all of you.

Thank you.

(this cheesy story was originally posted on my Patreon!)