A basic grounding in the principles and practices around artificial intelligence (AI), automation and cognitive systems is something which is likely to become increasingly valuable, regardless of your field of business, expertise or profession.
Fortunately, today you don’t have to take years out of your life studying at university to become familiar with this seemingly hugely complex technology. A growing number of online courses have sprung up in recent years covering everything from the basics to advanced implementation.
Some are aimed at people who want to dive straight into coding their own artificial neural networks, and understandably assume a certain level of technical ability. Others are useful for those who want to learn how this technology can be applied by anyone, regardless of prior technical expertise, to solving real-word problems.
In this post I will give a rundown of some of the best free ones which are available today.
Learn with Google AI
This newly launched resource is part of Google’s plan to broaden the understanding of AI among the general public. Material is slowly being added but it already contains a Machine Learning with TensorFlow (Google’s machine learning library) crash course.
The course covers the ground from a basic introduction to machine learning, to getting started with TensorFlow, to designing and training neural nets.
It is designed so that those with no prior knowledge of machine learning can jump in right at the start, those with some experience can pick or choose modules which interest them, while machine learning experts can use it as an introduction to TensorFlow.
Google – Machine Learning
This is a slightly more in-depth course from Google offered through Udacity. As such, it isn’t aimed at complete novices and assumes some previous experience of machine learning, to the point where you are at least familiar with supervised learning methods.
It focuses on deep learning, and the design of self-teaching systems that can learn from large, complex datasets.
The course is aimed at those looking to put machine learning, neural network technology to work as data analysts, data scientists or machine learning engineers as well as enterprising individuals wanting to make use of the plethora of open source libraries and materials available.
Stanford University – Machine Learning
This course is offered through Coursera and is taught by Andrew Ng, the founder of Google’s deep learning research unit, Google Brain, and head of AI for Baidu.
The entire course can be studied for free, although there is also the option of paying for certification which could certainly be useful if you plan to use your understanding of AI to increase your career prospects.
The course covers the spectrum of real-world machine learning implementations from speech recognition and enhancing web search, while going into technical depth with statistics topics such as linear regression, the backpropagation methods through which neural networks “learn”, and a Matlab tutorial – one of the most widely used programming languages for probability-based AI tools.
Columbia University – Machine Learning
This course is also available in its entirety for free online, with an option to pay for certification should you need it.
It promises to teach models, methods and applications for solving real-world problems using probabilistic and non-probabilistic methods as well as supervised and unsupervised learning.
To get the most out of the course you should expect to spend around eight to ten hours a week on the materials and exercises, over 12 weeks – but this is a free Ivy League-level education so you wouldn’t expect it to be a breeze.
It is offered through the non-profit edX online course provider, where it forms part of the Artificial Intelligence nanodegree.
Nvidia – Fundamentals of Deep Learning for Computer Vision
Computer vision is the AI sub-discipline of building computers which can “see” by processing visual information in the same way our brains do.
As well as the technical fundamentals, it covers how to identify situations or problems which can benefit from the application of machines capable of object recognition and image classification.
As a manufacturer of graphics processing units (GPUs), Nvidia unsurprisingly covers the crucial part these high-powered graphical engines, previously primarily aimed at displaying leading-edge images, has played in the widespread emergence of computer vision applications.
The final assessment covers building and deploying a neural net application, and while the entire course can be studied at your own pace, you should expect to spend around eight hours on the material.
MIT – Deep Learning for Self Driving Cars
As with the course above, MIT takes the approach of using one major real-world aspect of AI as a jumping-off point to explore the specific technologies involved.
The self-driving cars which are widely expected to become a part of our everyday lives rely on AI to make sense of all of the data hitting the vehicle’s array of sensors and safely navigate the roads. This involves teaching machines to interpret data from those sensors just as our own brains interpret signals from our eyes, ears and touch.
It covers the use of the MIT DeepTraffic simulator, which challenges students to teach a simulated car to drive as fast as possible along a busy road without colliding with other road users.
This is a course taught at the bricks ‘n’ mortar university for the first time last year, and all of the materials including lecture videos and exercises are available online – however you won’t be able to gain a certification.