Teaching myself Artificial Intelligence

The pandemic left me with time on my hands. I’m deciding to do something with it.

Hi I’m Shourya. I’m a computer science student at Surrey University, but I’m on an industrial placement this year. If you want to find out more about me, just click on the link below.

How I learnt to learn

While on placement, I’ve set myself a target of learning Artificial Intelligence in my spare time. I hope to have learnt it to a degree that lets me use it in some way for my final year project in September 2020.

Why am I doing this?

My one philosophy for learning is that if you truly have a vested interest in any area, studying it is no longer a chore.

Artificial intelligence has been around for decades. But with the increase of availability of high end computing units, and an unprecedented ease of access for novice developers such as myself, AI has become one of the fastest growing areas in technology in the past few years.

So I decided that it was going to be far more beneficial to be able to understand and learn a topic that’s growing more important with each passing day than to let technology pass me by.

My Goals

1. Learn enough to be dangerous. AI is an incredibly wide and incredibly complex area. At some point, I want to be able to understand a topic to the same extent as if I specialised in it for my degree.
2. Find uses for AI in my work. My aim is not just to study this area for the sake of its popularity. I want to be able to apply what I’ve learnt in the work I do currently on placement, and for it to have a significant role in my final year project.
3. Show others that everyone is capable of learning. By documenting my process on this page, I hope I can show to readers that everyone can (with the right amount of time and effort) achieve what they set out to do. And for those people that don’t know where to start, I hope this can give them the framework they need to pursue their interests in Artificial Intelligence.

My Curriculum

I’ve already had the fortune of being a computer science student for the past 2 and half years so I have programming experience under my belt. Because of this, some of the courses may not be the right starting point for you, and I will do my best to create the curriculum to be as beginner-friendly as possible.

Another thing to note is that my curriculum may not have the most robust structure. The majority of it is being made on the fly, and despite my hours of research into this, I may not have certain topics in the order recommended by an Artificial Intelligence PhD student.

Having already worked with some basic forms of AI in text and image analysis, I had some python skills under my belt. I’ve decided that my time is best spent building a strong foundation through theory, before applying into code.

It’s important to realise that I will be updating this course as time goes on, and I will aim to keep this course as accessible as possible. This means that everything shown below (other than the optional reading material) will be available for free.

Foundation Courses

To ensure the course is useable by people who’ve got little coding experience, these foundation courses are set up to work for everyone. Courses in bold are ones I am enrolled in.

• Udacity Programming Foundations with Python
• Team TreeHouse Python Track
• Udacity Intro to Data Science
• Intro to Python for Data Science Track

These courses give everyone a basic understanding of Data Science and programming in Python. The foundations built here will be incredibly useful when we attempt to explore AI in detail.

Advanced Courses

Once I feel that I have a good understanding of the basics of data science and python, I want to be able to apply and reinforce these basics onto new and more challenging topics. Because of this, I decided to enlist in Coursera’s state-of-the-art courses written by DeepLearning.ai and IBM.

• Introduction to Artificial Intelligence by IBM
• IBM Introduction to Machine Learning Specialization
• Deep Learning Specialization by deeplearning.ai

These courses are incredibly expansive, and contain numerous different topics and in many cases, real life examples of how to apply our knowledge of AI. My aim is to be able to complete all three by February, giving me plenty of time to start working on my own project from there on.

Reading Material

The greatest minds in the world read more books in a year than most of us do in a decade. To truly gain a mastery of this topic, I will need to go through some reading material. I have chosen the books below for their extensive coverage of the subject and in some cases due to their overwhelming popularity with the AI community, and I would urge everyone to at least read one of them if they feel they have the time.

• Naked Statistics by Charles Wheelan
• Artificial Intelligence: A Modern Approach by Peter Norvig and Steven Russell
• Deep Learning with Python by François Chollet
• Hands-On Machine Learning with Scikit-Learn and TensorFlow by Aurélien Géron
• The Hundred-Page Machine Learning Book by Andriy Burkov