Does someone need to know the inner workings of a combustion engine to learn how to drive a car?
Of course not.
You get in, you realise there’s a wheel, two or three pedals, a gear stick which you set to D (for drive) and then a collection of carefully designed signs to ensure you keep the car going the right way.
Once you’ve got the hang of things, it becomes second nature.
Then comes the race car driver. A race car driver stands to benefit if they know their car inside out. How fast it’ll go around the next corner, how much fuel is being pumped into the engine, whether or not the gear change between 3rd and 4th is worth it at 4,000 rpm. I’m making these up, I’ve never been a race car driver but that’s the things I imagine they think about.
The same thing goes for learning any other practical skill. Trying to learn everything before putting your knowledge to work is like trying to boil the ocean.
People have asked in the past, how did you learn all the math behind machine learning?
I didn’t. I still haven’t. I barely know all the code, all the models, all the ways to prepare data.
Instead, I learn what I need to learn when I need to learn. It’s much easier for me to learn something when I can see the roadblock it’s removing. It turns learning into a game, into a playful dance. I jump from one thing to the next. Treating every day as day one.
The same goes with anything else. A practitioner learns through trial and error. They setup an experiment and see if it works. If it doesn’t, they know where not to go. And if it does, great.
Following your curiosity and learning what you need to learn when you need it doesn’t mean it won’t be difficult. Learning by definition is hard. Have you ever tried to make new brain cells appear on demand? Or connected multiple different brain cells? How’d it go? Because that’s what’s happening when you learn.
The next time you get down for not knowing everything, remind yourself, boiling a kettle is easier than boiling the ocean.
Hat tip to George for telling me the saying. He said it on a whim one day and I’ve held onto it ever since. You never know when these things are going to come in handy.
PS the YouTube comment is from my latest livestream answering as many questions as I could about the new machine learning workshop my friend and Andrei and I just released on Udemy. It’s a beginner friendly approach to teaching machine learning, if you’re interested, there’s a promotion going for the rest of January.