'Simple' Solution to 'Complex' (Analytical) problems

Let me tell you a story of a personal experience. A few days back, I was solving a puzzle. It was a 3D puzzle with 12 wooden pieces with some sunken and extended portions that would fit to make a structure like following:

Spoiler Alert: If you aspire to ever solve this 'complex' puzzle yourself, you should stop reading now. Else, I'll be revealing a 'simple' solution here on.

Now, as typically one would start, I started arranging the pieces one at a time, hoping it would all come together by itself. But the puzzle, as you can imagine, absolutely didn't oblige. Now, the neural network in our fantastic brains is capable to pickup, learn & solve even such complex puzzles in minutes, if not seconds, almost intuitively- repeat till you succeed - little better every time - mode. However, I decided to go back to basics. I thought, let me see through the pieces properly. If the 12 pieces of wood were 12 rows of data, then, this step would be the Exploratory Data Analysis. Its getting so familiar with your data that you could say, "Hey Data, Now I know you, as much as you do yourself."

Bear with me, for the 3 steps of my solution. The learnings come AFTER these 3 steps.

My simple analysis revealed that:

Ok, now, this meant that I needed to make 3 squares that would fit in 3D. This step is finding patterns in the data- Data mining. Again, a simple analysis revealed that - clearly not all ways how pieces could fit each other, would be the correct way to fit them.

Then, I took the right patterns, and fit them one over the other, put some brute force towards the last pieces to make them go and sit in the right place and completed the puzzle. I thanked that the puzzle was in soft wood, else, it would not have adjusted to fit in.

However, that's not the end. Here comes the interesting bit. How do I take it off again, to store the puzzle back in its box, given the pieces had fit in with some brute force, remember?

To unpack the puzzle, I just picked it up from one corner and gave it a mild shake. Can you imagine what happened? The entire structure started to come off on its own. But how did something that took force to complete, come down without any force? That's because, nature is a far far smarter data scientist. In real time, with each shake it tests and learns about each joint and leverages the weakest joint each time and unravels the puzzle. It is like knowing your data to the deepest extent, and with every possible interaction, so that you know every kink, every corner. You know, they are not 5 white pieces, but each has its own character and faults, and that too when in interaction with some other piece. In fact, not even trying to learn, you figure that you just know it when you need to. Going back to the learning bit as well. How nature (our brain learns) is so much more efficient. When you think about it, its so simple, that makes it complex to believe.