Introduction
When the world’s smartest companies such as Microsoft, Google, Alphabet Inc., and Baidu are investing heavily in Artificial Intelligence (AI), the world is going to sit up and take notice. Chinese Internet giant Baidu spent USD1.5 billion on research and development.
And as proof of China’s strong focus on AI and Machine Learning, Sinovation Ventures, a venture capital firm, invested USD0.1 billion in “25 AI-related startups” in the last three years in China and the U.S.
Research shows that although genuine intelligence may still be a bit far off, AI and Machine Learning technologies are still expected to reign in 2017. Try reading up on Microsoft Project Oxford, IBM Watson, Google Deep Mind, and Baidu Minwa, and you’ll understand what I am trying to get at.
In 2015, Gartner’s Hype Cycle for Emerging Technologies introduced Machine Learning (ML), and the graph showed (Figure 1) that it would reach a plateau in 2 to 5 years. Big players such as Facebook and Amazon are increasingly exploiting the advantages of this concept, which is derived from artificial intelligence and statistics, to extract meaning from huge amounts of (big) data.
Research predicts that the AI market will grow to about USD37 billion by 2025; in 2015 it was about USD645 million!
Source: Gartner
Difference between Machine Learning and Artificial Intelligence
Artificial Intelligence (AI) and ML are not interchangeable terms. ML is sort of a subset of AI, which is a part of computer science trying to develop “machines capable of intelligent behavior.” Then, what is Machine Learning (ML)? “The science of getting computers to act without being explicitly programmed,” says Stanford. So you get that difference? You need both AI and ML experts to make smart machines that are truly intelligent.
Why are Machine Learning and Artificial Intelligence “Hot”?
"Machine learning is a core, transformative way by which we’re rethinking everything we’re doing” — Sundar Pichai, Google CEO
The pervasive commercial success of machine learning/artificial intelligence is visible everywhere—from Amazon recommending what movies you might like to see to self-driving Google cars that can tell a tree from a pedestrian.
AI/ML has changed how data-driven business leaders make decisions, gage their businesses, study human behavior, and view predictive analytics. If your organization needs to unleash the benefits of this extraordinary field, you need the right minds—quants and translators.
With breakthroughs such as parallel computation that’s cheap, Big Data, and improved algorithms, utilitarian AI is what the world is moving toward. The increased need to handle huge amounts of data and the number of IoT connected devices that define the world today reinforce the importance of machine learning.
AI/ML, with tons of potential, is a great career choice for engineers or data mining/ pattern recognition enthusiasts out there. Also, Machine Learning is integral to data science, which is touted as the sexiest job of the 21st century by the Harvard Business Review.
An Evans Data Corp. study found that 36% of the 500 developers surveyed use elements of ML in their Big Data or other analytical projects. CEO Janel Garvin said, “Machine learning includes many techniques that are rapidly being adopted at this time and the developers who already work with Big Data and advanced analytics are in an excellent position to lead the way.”
She added: “We are seeing more and more interest from developers in all forms of cognitive computing, including pattern recognition, natural language recognition, and neural networks and we fully expect that the programs of tomorrow are going to be based on these nascent technologies of today.”
So, for people who have a degree in Computer Science, Machine Learning, Operational Research, or Statistics, the world could well be their oyster for some time to come, right?
List of Courses
I’ve put together (and agonized a bit over what to add and what not to) a few free top ML and AI courses that will help you become the next ML expert Google or Apple hires. Of course, it is hard work, but if you are willing to pursue something, you’ll discover ways like these to succeed.
Machine Learning Courses
1. Machine Learning by Andrew Ng
Co-founder of Coursera, Andrew Ng, takes this 11-week course. He is an Associate Professor at Stanford University and the Chief Scientist at Baidu. As an applied machine learning class, it talks about the best machine learning techniques and statistical pattern recognition, and teaches you how to implement learning algorithms.
Broadly, it covers supervised and unsupervised learning, linear and logistic regression, regularization, and Naïve Bayes. He uses Octave and MatLab. The course is rich in case studies and recent practical applications. Students are expected to know the basics of probability, linear algebra, and computer science. The course has rave reviews from the users.
Go to Course: Start learning
2. Udacity’s Intro to Machine Learning
A part of Udacity’s Data Analyst Nanodegree, this approximately 10-week course teaches all you need to know to handle data sets using machine learning techniques to extract useful insights. Instructors Sebastian Thrun and Katie Malone will expect the beginners to know basic statistical concepts and Python.
This course teaches you everything from clustering to decision trees, from ML algorithms such as Adaboost to SVMs. People also recommend you take the foundational Intro to Data Science course which deals with Data Manipulation, Data Analysis, Data Communication with Information Visualization, and Data at Scale.
Go to Course: Start learning
3. EdX’s Learning from Data (Introductory Machine Learning)
Yaser S. Abu-Mostafa, Professor of Electrical Engineering and Computer Science at the California Institute of Technology, will teach you the basic theoretical principles, algorithms, and applications of Machine Learning.
The course requires an effort of 10 to 20 hours per week and lasts 10 weeks. They have another 5-week-course, Machine Learning for Data Science and Analytics, where newbies can learn more about algorithms.
Go to Course: Start learning
4. Statistical Machine Learning
Your instructor of the series of video lectures (on YouTube) in Advanced Machine Learning is Larry Wasserman, Professor in the Department of Statistics and in the Machine Learning Department at the Carnegie Mellon University.
The prerequisites for this course are his lectures on Intermediate Statistics and Machine Learning (10-715) intended for PhD students. If you can’t access these courses, you need to ensure you have the required math, computer science, and stats skills.
Go to Course: Start learning
5. Coursera’s Neural Networks for Machine Learning
Emeritus Distinguished Professor Gregory Hinton, who also works at Google’s Mountain View facility, from the University of Toronto teaches this 16-week advanced course offered by Coursera.
A pioneer in the field of deep learning, Hinton’s lecture videos on YouTube talk about the application of neural networks in image segmentation, human motion, modeling language, speech and object recognition, and so on. Students are expected to be comfortable with calculus and have requisite experience in Python programming.
Go to Course: Start learning
6. Google’s Deep Learning
Udacity offers this amazing free course which “takes machine learning to the next level.” Google’s 3-month course is not for beginners. It talks about the motivation for deep learning, deep neural networks, convolutional networks, and deep models for text and sequences.
Course leads Vincent Vanhoucke and Arpan Chakraborty expect the learners to have programming experience in Python and some GitHub experience and to know the basic concepts of ML and statistics, linear algebra, and calculus. The TensorFlow (Google’s own deep learning library) course has an added advantage of being self-paced.
Go to Course: Start learning
7. Kaggle R Tutorial on Machine Learning
DataCamp offers this interactive learning experience that’ll help you ace competitions. They also have an Introduction to R course for free.
Go to Course: Start learning
8. EdX’s Principles of Machine Learning
A part of the Microsoft Professional Program Certificate in Data Science, this 6-week course is an intermediate level one. It teaches you how to build and work with machine learning models using Python, R, and Azure Machine Learning.
Instructors, Dr. Steve Elston and Cynthia Rudin talk about classification, regression in machine learning, supervised models, non-linear modeling, clustering, and recommender systems. To add a verified certificate, you’ll need to pay.
9. Coursera’s Machine Learning Specialization
The University of Washington has created five courses, with practical case studies, to teach you the basics of Machine Learning. This 6-week course which requires between 5 and 8 hours of study a week, will cover ML foundations, classification, clustering, regression, recommender systems and dimensionality reduction, and project using deep learning.
Amazon’s Emily Fox and Carlos Guestrin are the instructors, and they expect the learners to have basic math and programming skills along with a working knowledge of Python. Course access is free though getting a valid certificate is not.
Go to Course: Start learning
Artificial Intelligence Courses
1. EdX's Artificial Intelligence
This exciting course from EdX talks about AI applications such as Robotics and NLP, machine learning (branch of AI) algorithms, data structures, games, and constraint satisfaction problems. It lasts 12 weeks and is an advanced-level tutorial from Columbia University.
Go to Course: Start learning
2. Udacity’s Intro to Artificial Intelligence
The course is expected to teach you AI’s “representative applications.” It is a part of its Machine Learning Engineer Nanodegree Program. Instructors Sebastian Thrun and Peter Norvig will take you through the fundamentals of AI, which include Bayes networks, statistics, and machine learning, and AI applications such as NLP, robotics, and image processing. Students are expected to know linear algebra and probability theory.
Go to Course: Start learning
3. Artificial Intelligence: Principles and Techniques
This Stanford course talks about how AI uses math tools to deal with complex problems such as machine translation, speech and face recognition, and autonomous driving. You can access the comprehensive lecture outline—machine learning concepts; tree search, dynamic programming, heuristics; game playing; Markov decision processes; constraint satisfaction problems; Bayesian networks; and logic— and assignments.
Go to Course: Start learning
4. Udacity's Artificial Intelligence for Robotics by Georgia Tech
Offered by Udacity, this course talks about programming a robotic car the way Stanford and Google do it. It is a part of the Deep Learning Nanodegree Foundation course. Sebastian Thrun will talk about localization, Kalman and Particle filters, PID control, and SLAM. Strong grasp of math concepts such as linear algebra and probability, knowledge of Python, and programming experience are good-to-have skills.
Go to Course: Start learning
Summary
In this post, a few of the listed courses are meant to help you get started in the exciting and fast-growing field of Machine Learning and Artificial Intelligence. Others take you through slightly more advanced aspects. The courses listed are free and the only thing stopping you from getting the most out of them will be a lack of commitment.
These world-class courses, which focus on a specific area of learning, are great stepping stones to lucrative and amazing careers in machine learning, data science, and so much more. If you don’t want the Baxters of the world to make you obsolete, you best teach them just who the master is.
So once you identify your learning goals, and assuming you have reliable access to technological requirements, be self-disciplined, build a study plan, set time limits, stay on schedule, work effectively with others, and, most of all, find ways to stay motivated.
