AI as Infant: The Layman’s Guide to Neural Networks

Jane Doerflinger
Feb 3, 2019 · 5 min read

As AI becomes more ubiquitous, many of us are scrambling to learn to code and master mathematics. Your colleagues may have started talking about potential AI initiatives. Maybe, late at night, your vague technological FOMO leaves you scouring wikipedia.

But AI are made in our image. So rather than looking outward, we can understand them best by looking at ourselves.

With that in mind, let me introduce you to the most exciting form of AI: the neural network.

Image for post
Image for post

The apt NN/baby comparison isn’t coincidental. As the name suggests, NNs are structured like human brains. And like us, they start out as blank slates, gaining understanding as they’re exposed to more data.

Given the fundamental similarities, they also share many quirks, strengths, and downfalls with babies. For example, they’re….

Hard to Understand

If your baby has decided to shriek, you probably don’t know why. You may ask, but it’s not going to get you anywhere.

Similarly, even compared to other algorithms, NNs can’t explain their reasoning. You just have to look at the output, look at the input, and guess why it’s doing what it’s doing. This is known as the black box problem.

Prone to Mistakes When They See New Things

If a baby has only ever seen dogs, when she sees her first cat, she’s likely to call it a “doggie.”

NNs often make similar mistakes when their datasets aren’t carefully designed. Be sure to include edge cases — hairless cats, cats with 3 legs, etc.

Always Getting Into Trouble

Kids are always tugging down lamps, scattering Tupperware, and picking up curse words.

NNs can be mischievous too. Once in a while, when they’re supposed to practice English, they invent a new language. When you want them to transform and revert images, they cheat by encoding the original into the elaborations.

Incredibly Flexible

Humans aren’t built for reading. Nothing in our neurology or our evolutionary past suggests we should be able to do it. But kids learn it anyway.

Similarly, NNs are also able to process all forms of data — images, numbers, video, text, and audio — instead of specializing in just one or two of them.

Needy and Expensive

NNs require larger datasets and more computing power than other models. So if another algorithm will get the job done, don’t build a NN. Be like your aunt who skipped kids to travel the world.

Completely Without Baggage

Kids’ ignorance can sometimes lead to a kind of genius. They rename vultures “flamingo witches,” the day before yesterday “lasterday,” and armoires “clothes fridges.” If you challenged world-class writers to rename things, I don’t think they’d do better.

AI and ML models tend to have a similar talent. Without the weight of experience and cultural expectations, they can create surprising solutions that would never have occurred to an expert. One fascinating example are the ugly moves of chessbots. They fly in the face of traditional chess theory. And they win games.

So what do NNs do?

If a one year old does it, a NN can too.

  • Understand language? Check.
  • Recognize objects? Check.
  • Interpret facial expressions? Check.
  • Imagine familiar objects? Check.

By the same token, one year olds and NNs struggle with many of the same tasks.

Image for post
Image for post
Image for post
Image for post

But what about some of the more extraordinary things NNs can do? Like learn chess, identify skin cancer, and generate photorealistic images?

You got me. I admit there are some differences between babies and NNs.

  • Babies’ brains need to translate electrical signals into actions in the outside world — like eating cheerios or saying the word “banana.” But, generally, NNs operate on their own turf. Without that layer of translation, a lot of tasks become much easier.
  • NNs are focused and obedient. If you tell one to practice chess for 10 hours, when you check on it again, it will have mastered the game. If you try the same thing with a baby, it will have mastered chewing on chess pieces and looking out windows at birds.

So imagine a super-focused, super-obedient, disembodied baby. I believe that one could achieve any of the tasks NNs can. Babies’ brains aren’t limited by simplicity or a lack of processing power. (Far from it.)

What does this mean for me?

You live alongside the most bizarre, amazing thing that humanity has ever created. Tireless helpers are ready to assist you with an ever-growing range of tasks. It’s up to you to tap into that potential.

If you want to put one to work in your company, look for an AI API, which provides a pre-trained model for pay-as-you-go use. For super simple, low-investment projects, go with one optimized for one task — like conversation or image recognition — rather than something more general. (These are all marketed as NNs, but rumor has it many of them are not. Which isn’t a bad thing. Just don’t be surprised if they work a little differently than you expect.)

If you want a quick sense of NNs’ nuts and bolts, check out this video.

And if you want to get deep into the weeds, try building your own NN in Python. While the mechanics of NNs can be tricky, the code itself is shockingly simple. If want to learn to code, and you’re interested in data analysis, Python is the perfect place to start.

If you get stuck with any of these, don’t hesitate to ask StackOverflow and for help. It takes a village, afterall.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store