Deep Learning for total beginners

How to make sense of neural networks if you don’t have technical background?

Przemek Chojecki
Nov 17, 2019 · 3 min read

When we use our computers to do something, it has become an accepted practice to analyze the results, then decide what’s working and what’s not. However, with deep learning, researchers are able to use what they’ve learned in order to identify patterns, trends, and predict future patterns.

Neural networks are used for facial recognition

In order to do this, researchers have been using an approach called feature engineering. Using this approach, researchers build models from many different inputs and then use those features to predict the next input to which they will likely be applied.

What is Deep Learning

Deep Learning is a new kind of model which is built from many different models, and it’s designed specifically to learn things by itself, and then learn a new thing, with an incredible efficiency in that sense.

We have a lot of deep learning models like the convolutional layers. We use this for facial recognition. We also use these neural network models like backpropagation and feed-forward neural networks. These are things where we have lots of inputs and lots of outputs, but they’re all labeled. They’re all labeled with different kinds of properties, and these neural networks are actually learning how to classify images. They’re learning how to distinguish between different kinds of objects.

But deep learning is an algorithm that’s specifically designed to learn, so it’s learning by itself. It has no labels, and it does something very different than what we’re normally talking about. That’s why it’s called “deep learning,” as in it learns by itself. And I would say that it’s a more important innovation than the internet. I would say that the internet is less important than deep learning because the internet was really about getting information. Deep learning is getting information and making it easier to get.

Algorithmic approach to life

Deep learning is all about the algorithms. There’s more and more that are coming out with more and more of this amazing research. So the deep learning revolution is happening right now.

It is a huge innovation and I believe that the deep learning revolution has the potential to really transform the future. I believe that it has the potential to fundamentally change how we are doing business, how we are doing things, how we build things, and the way we think about things.

One thing that’s amazing is when you take deep learning and you use deep learning to improve any system, whether it’s a computer whether it’s a robot, whether it’s a driverless car, you can improve them significantly or do it at all.

When I talk about deep learning and machine learning, I always say you have to go deep. And I mean, deep in the sense that, in order for the machine to make sense of this information, to have the knowledge that it needs to make a decision, you have to go deeper in data and algorithms and layers.. And the deeper you go, the better you get.

Artificial Intelligence Hype

And there’s always been some fear that, oh, well, you know, machine learning is so complicated. Maybe you don’t understand how to go deep enough. Or, maybe, you know, it’s a tough problem to solve. And that’s why the recent hype about deep learning was so much fun.

I love hype because it’s all hype. You can’t predict the future. You can’t predict how it’s going to turn out. But you have to tell people it’s going to happen. And the fact that we’re now at the point where we’re using it for image recognition and speech recognition is only making me more hopeful about the future.

Embrace the changes!

Przemek Chojecki

Written by

AI entrepreneur with a PhD in mathematics, Forbes 30 under 30. My Data Science Job book is out: https://amzn.to/31jXMAS

Analytics Vidhya

Analytics Vidhya is a community of Analytics and Data Science professionals. We are building the next-gen data science ecosystem https://www.analyticsvidhya.com

More From Medium

More from Analytics Vidhya

More from Analytics Vidhya

More from Analytics Vidhya

The Illustrated Word2vec

More from Analytics Vidhya

More from Analytics Vidhya

Financial Transaction Fraud Detection

219

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