Don’t have time to read? Then give it a listen.
In this modern era, Artificial Intelligence (AI) and Machine Learning (ML) are the two hot buzzwords. Though these two are often used interchangeably, they’re not quite the same thing. Artificial Intelligence is the intelligence demonstrated by the machines being able to carry out the tasks in a way that we would consider smart. Whereas Machine Learning is a subset of artificial intelligence where instead of programming a computer, we give it the ability to learn with data. And From detecting skin cancer to sorting cucumbers, Machine Learning has granted computer systems a new ability!
The idea of machine learning is not new. The machine learning algorithms have been around since 1760s. In fact the first machine learning algorithm was discovered on 1763 from by Thomas Bayes. The work was a Bayesian network and it was published 2 years after his death.
But why the sudden hype? The problem with machine learning was that it requires huge amount of data and computational power to work which was definitely not available for general people in the past centuries. Big companies like Google and Facebook have been implementing these algorithms for the past decades. But their hardware is not feasible for everyone. However with the advent of cheap and powerful CPUs and Graphical Processing Units(GPUs), machine learning is more available to public than ever.
Machine learning applications can mainly be seen on Image Processing(object detection, classification) and financial prediction. However they are not limited to these fields only. Social Science, material science, geography and even medicine have found their application of machine learning. So, ostensibly machine learning may look like a skill limited for computer programmers but it is not. It has become a necessity for workers and researcher from all types of field. As Eric Scmidt has described Machine Learning as the next transformation of technology.
Learning machine learning and applying it on their respective field is a monumental task. Specially for those who do not come from programming background. Also the vast amount of available resources can sometimes be confusing. And to make it worse most of the resources assumes your knowledge in programming. In this article some of the best resources and prerequisites will be explored.
As machine learning is implemented on computers, it would definitely require programming skills. These algorithms can be implemented on any popular programming languages like C, C++, JAVA, Python, JULIA and R.
For beginners Python is the preferred language. Because it is the most human readable language and easy to learn. Users can focus on machine learning rather than exhausting syntax and data structure errors.
Along with Programming language some fundamentals in math is also required. Learners should be familiar with topics like statistics, linear algebra, matrix and calculus. Learners need not have deep understanding in these topics as ultimately computer can do most of the job with built in libraries. However it is imperative to have basic understanding of these to know what is actually going on under the hood!
Prerequisites in a nutshell:
- Basic Understanding of Linear Algebra
- Basic Understanding of Probability
- Fundamental of knowledge of a relevant programming language
Now, as we’ve learnt about some of the prerequisites of machine learning, it’s time to look into some of the best resources for starting machine learning at home! The resources discussed in this article are mainly based on Python because of its simplicity for beginners. Let’s start then.
This site is an excellent place to begin machine learning. With some basic understanding in python, learner can jump into the machine learning course. This site focuses on the basics of algorithms. Instead of using the built in machine learning libraries like scikit-Learn and tensorflow, it focuses on implementing algorithms from the scratch. Also many practical examples are shown along the way.
This Coursera course is another excellent place for beginners released in 2011. Like the previous site this also focuses on algorithms.This course starts steadily by building the mathematical foundation. However the implementations and evaluation are shown in MATLAB which is also taught along the course. This is by far the highest rated course on Machine Learning. The instructor, Andrew NG, is the founder of Coursera and teaches the popular CS 229 Machine Learning Course at Stanford.
Another coursera course. This is the modern version of the course by Stanford University. This too starts from the fundamentals. However focus of this course is deep learning algorithms, an advanced section of machine learning.
A sub site of Medium. This site provides some good reads on fundamentals and implementations for self paced learners.
Also, Books by Packt Publishing are also some good resources.
There are tons of other resources out there like these. If the courses do not match to the learners liking, they can switch to other options whenever they want. The important thing is to start. And once you did it, you’ll be fascinated for sure!
Co-authored by Wasima Noor Iqra
This article’s audiobook is read by Sadia Raisa
Visit 10 Minute School’s website: www.10minuteschool.com/
If you want to unleash the writer in yourself and contribute for 10 Minute School, simply send your write-up to the following e-mail address: email@example.com