Here is a list of self-study options for machine learning practitioners to get started with the basic machine learning theory and termiology.
Data Science for Beginners. Get a quick introduction to data science in five short videos hosted by Microsoft data scientist Brandon Rohrer.
Machine Learning Is Fun!. This guide is for anyone who is curious about machine learning but has no idea where to start.
Practical Data Science with the Cortana Intelligence Suite. During this course, you learn about analytics workload, the Cortana Analytics process, the foundations of data transfer and storage, data source documentation, storage and processing using Machine Learning, and other techniques.
edX / Microsoft: Data Science and Machine Learning Essentials. Learn key concepts of data science and machine learning with examples on how to build a cloud data science solution with R, Python and Azure Machine Learning from the Cortana Analytics Suite.
edX / Microsoft: Principles of Machine Learning. Get hands-on experience building and deriving insights from machine learning models using R, Python, and Azure Machine Learning.
edX / Microsoft: Implementing Predictive Solutions with Spark in Azure HDInsight. Learn how to use Hadoop technologies in Microsoft Azure HDInsight to create predictive analytics and machine learning solutions.
University of Washington / Dato: Build Intelligent Applications. Master machine learning fundamentals in five hands-on courses.
Stanford: CS229 – Machine Learning (Andrew Ng). Deep enough to understand the basics very well, while not overwhelming with theoretical details. You need to be pretty comfortable with linear algebra and vector calculus. The exercises here are incredibly valuable.
Stanford/Coursera: Machine Learning (Andrew Ng). Solid, practical introduction to ML. Teaches core concepts but doesn’t give a good sense of all that’s required to do real-world work. Math-heavy, uses Octave (a Matlab clone), and recommended for engineers and others wanting to get started with ML.
Caltech: Machine Learning (Yaser Abu-Mostafa). Introductory but less friendly for novices. Goes into most of the same topics as Ng’s course, but more deeply and has a more theoretical bent. Recommended for students and practicing data scientists.
Georgia Tech/Udacity: Machine Learning: Supervised, Unsupervised, Reinforcement. Introductory and similar coverage as the Caltech class.
Stanford: Introduction to Statistical Learning (Trevor Hastie and Rob Tibshirani). Equations for all topics well explained in plain English, with pointers to specific chapters in the textbook for a deeper mathematical explanation. Each chapter has a lab with examples in R. Book available for download.
Johns Hopkins/Coursera: Data Science. A sequence of courses to learn to be a data scientist and apply your skills in a capstone project.
Eindhoven University of Technology/Coursera: Process Mining — Data Science in Action. The course explains the key analysis techniques in process mining. Participants will learn various process discovery algorithms.
Harvard: Data Science. Introduces methods for five key facets of an investigation: data wrangling, cleaning, and sampling; data management; exploratory data analysis; prediction; and communication of results.
University of Washington: Certificate in Data Science. Become equipped with the fundamental tools, techniques and practical experience to acquire valuable insights from data sets at any scale — from gigabytes to petabytes.
University of Washington: Masters of Science in Data Science. An interdisciplinary curriculum developed through a collaboration among six nationally recognized departments and schools at the UW, with the input of top companies.
Udacity: Data Analyst Nanodegree. Extract, transform, and load data; exploratory data analysis; classify data and use applied statistics and ML algorithms; and communicate data analysis.
IBM: Big Data University. Big Data University is an IBM initiative to spread big data literacy.