June 2nd 2020
An AI enthusiast, who is optimistic as well as pessimistic about future.
“Anybody can code” , I know this sentence sounds cliche so let me give you another one “Anybody can learn AI”. Well, know it sounds overwhelming except if you are not a PhD or a mad scientist.
But believe me, at the end of this article irrespective of you being a programmer/non-programmer you will get to know the path and the best resources to learn AI in a sequential manner.
Before starting I just want to make it clear that AI IS A VAST FIELD which itself comprises of sub-fields like Machine Learning, Deep Learning, Reinforcement Learning, Natural Language Processing, and many more. To be honest AI is like an ocean and its sub-fields are like seas so learning AI completely is impossible unless you are a SUPERHUMAN AI.
So after enough trash talk, let’s START
If you are a non-programmer who is completely new to programming or a person who thinks that programming is hard or it’s not for you then this course is the best place to start. It will cover the fundamentals or heart of programming and Computer Science. This course is taught by Prof. David J. Malan who is super dynamic and energetic in nature. If you don’t have the time to watch the entire course then at least watch the first lecture. If you already know the fundamentals of programming then you can skip this course.
The above course is 4 ½ hours long which will cover topics like Installation, variables, strings, list, tuples, functions, object-oriented programming concepts, and a lot more. By the way, this course is practical in nature. If you already know Python Programming then you can skip this course.
Now, this is the part where your AI journey will truly start. This course is theoretical in nature and is taught by the most renowned guy in the entire AI & Machine Learning industry, Prof. Andrew Ng. I mean anyone who has some knowledge in the field of AI mostly knows him. The best part about this course is that it’s short, concise, interesting & can be understood by anyone who has or hasn’t any knowledge about AI or programming in general.
This course beautifully answers some common but interesting questions like
- What is AI & Machine Learning?
- What makes an AI company?
- What Machine Learning can or cannot do?
- How to choose a Machine Learning or an AI project?
- What is the workflow of a Machine Learning project?
- Discrimination/Bias, Adversarial attacks on AI,
- AI and jobs
and many more.
One of the important question which will mostly come into your mind while learning AI & Machine Learning is:
Should we know the internal working of the algorithms in-depth or learn it superficially by implementing it with the help of the existing machine learning or deep learning platform
I would suggest you, to learn the inner working of all the algorithms before implementing them using any external libraries, but in the end the decision is up to you.
It’s one of the most popular course out there for Machine Learning which is taught by Prof. Andrew Ng. More than 3.1 million people have already enrolled in this course at the time of writing.
The best part about this course is that it’s one of the most in-depth courses out there for Machine Learning which dares to teach the inner working of the algorithms and the math’s behind it.
This course uses Octave/Matlab for coding the algorithms, but I would highly recommend you to code those algorithms using Python as it’s an industry standard.
So, after you have implemented those algorithms from scratch in the previous course or you are not comfortable with math’s related to Machine Learning then this is the right course for you.
This course is the best and the most underrated one found on YouTube. The best part about this course is the easy to understand explanation. In this course libraries like numpy, pandas, matplotlib, and sklearn are used to implement and visualize various Machine Learning algorithms. Using external libraries like the above ones you can easily implement these algorithms within a few lines of code.
Apart from covering all the standard Machine Learning algorithms like Linear Regression, Logistic Regression, Decision Tree, Random Forest, Support Vector Machine(SVM), K Means Clustering & Naive Bayes it also covers various topics like Gradient Descent, Saving Models, Dummy Variable & One Hot Encoding, etc.
So, after you are comfortable with Machine Learning now is the time to learn Deep Learning which is a sub-field of Machine Learning. Deep Learning algorithms are the one that drives the recommendation and personalization system of Netflix, Amazon, YouTube, and a tonne of large corporations, startups.
This is a specialization course in Coursera which is taught by Prof. Andrew Ng and it comprises 5 courses.
- Neural Network and Deep Learning
- Improving Deep Neural Networks: Hyper-parameter tuning, Regularization, and Optimization
- Structuring Machine Learning Projects
- Convolutional Neural Network
- Sequence Models
The best part about this specialization is its in-depth material, lectures which cover the technical and mathematical parts about the algorithms and techniques. Python is used for coding those algorithms.
Tensorflow & Pytorch are the open-source deep learning frameworks which are dominating the entire field of AI. Tensorflow is backed by Google and Pytorch is backed by Facebook.
On the basis of popularity and downloads Tensorflow has the upper hand but when it comes to the AI research community they strongly back Pytorch. So it’s up to you which framework or platform you want to choose. I would recommend to try both of them and see what fits you the best.
Like other specialization courses on Coursera, it comprises of 4 courses.
- Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning
- Convolutional Neural Networks in TensorFlow
- Natural Language Processing in TensorFlow
- Sequences, Time Series and Prediction
These courses are easy to grasp and are pretty practical in nature. It’s taught by the Instructor Laurence Moroney who is an AI Advocate at Google. Here you will build practical applications using Tensorflow.
If you want to learn Pytorch framework then there is no better place then there own website. Their tutorials are divided into small chunks that cover most of the fundamental parts of Pytorch and how to implement different Deep Learning modules using it.
To be honest, going through all of the above courses is going to take a lot of time so it depends on you and your needs what you want to prioritize. At last, I just want to say that “Anybody can learn AI and Everybody should learn AI because AI is the next Industrial Revolution“.