Introduction To A High Schooler’s Guide To Deep Learning and AI

Ada Tur
4 min readJul 24, 2020

Deep learning is a scary subject — at least that’s how it seems to a lot of people who don’t really understand the concept. See, deep learning and artificial intelligence has been one of the reasons why so many problems these days are being solved much quicker and better than before, and because of that, so many people are interested, but also scared of trying to learn it. Most of the time, those who are interested will look up some course or tutorial on how to learn how to code for artificial intelligence, but they will soon realize that most of these tutorials require a basic understanding of the field, or that the concepts are far too complex for their understanding. Furthermore, most high schoolers don’t have the math foundation that typical classes expect, which can be overwhelming once they see the immense amount of math. At least, that’s how I felt when I began to learn.

But, once you understand the concepts, artificial intelligence allows you to do a variety of incredible, different things. From creating image classifiers to generating poetry to creating solutions to medical problems, these are the projects that excite creators and scientists and these are the projects that inspire people like you, the reader, to begin learning how to code with artificial intelligence. And my job here is to break down all of the concepts you need to know so that no matter your experience or knowledge, you will be able to understand and replicate these ideas to write your own, new creations.

Firstly, I feel that you all deserve a brief introduction to the person teaching you everything that you will be seeing in the near future. I am a high schooler from California who has been learning and working with AI for over a year, and I have worked with companies on technology using artificial intelligent systems and founded a school club teaching other high schoolers deep learning, similar to what these articles will be about. With the knowledge I have, I feel that I am not only able to teach beginners artificial intelligence, but I am also able to simplify these concepts down so that anyone can understand and succeed. So why don’t we get started?

“Artificial intelligence is the new electricity” — Andrew Ng


Learning Python is essential for doing most tasks with AI. Python is a relatively simple and easy to understand language, and because most libraries are supported by Python, it has become the main language for deep learning. Having a bit of understanding will make your time reading through my posts much easier.


While there are ways that you can completely avoid doing most of the math in AI, it is incredibly crucial that you understand the math behind certain things so that you can understand how they work and how to improve them. Even though your libraries will do most of the work, any good scientist knows everything going on behind the scenes, and this requires some knowledge in Algebra, Calculus, and matrices and vectors. I assure you that you will be able to understand every calculation, definition, formula, and procedure done in this entire series as I explain.


The libraries that we will be using for this course are Numpy and PyTorch. Numpy is a library that offers a large amount of mathematical operations that makes our job much easier, and PyTorch, similar to Keras and Tensorflow, is a library that gives us machine learning functions that will be crucial for our work. We will likely be using some other libraries occasionally, but Numpy and PyTorch will be our main ones.

So…What Is AI?

To put it simply, artificial intelligence is how a computer can, once given a task, complete the task while also improving its chances of success. For example, one of the earliest tasks that we will go over is linear regression, where the computer, once given a set of points or data, can create a line of best fit through the points and make it more and more accurate as it analyzes the points.

Source: Aunalytics

You may have also heard of the terms ‘machine learning’ and ‘deep learning’, and most people don’t understand how the three are different. However, it’s rather simple; deep learning is a type of machine learning which is a subset of artificial intelligence.

Artificial intelligence has hundreds of purposes and uses in almost every existing field today. It has been used to solve problems in medicine, biology, the stock market, language processing and virtual assistants, image classification, speech and video comprehension and creation, computer vision, and more. People have even been able to create AI’s that can play games, as you may have seen in player v.s. computer chess games and whatnot.

“Machine intelligence is the last invention that humanity will ever need to make” — Nick Bostrom

All of this may seem daunting, but I assure you that it is much simpler than it seems. You will soon be able to understand everything I have gone over and you will have had experience doing projects, practice, and more as we go along in this course. In the next few lessons, I will be going over the basic math that is needed for you to understand everything. Then, I will explain linear and logistic regression and you will be able to do some quick projects that will be fun, simple, and explanatory. I hope to see you in the next article soon!



Ada Tur

Undergraduate student at McGill University studying Computer Science and Linguistics. Interested in NLP, deep learning, and Creative AI