Machine Learning (6.867)
MIT's graduate-level introduction to machine learning covers classification, linear regression, boosting, support vector machines, hidden Markov models, and Bayesian networks. The course builds intuition for these methods alongside a formal treatment of statistical inference, the underlying theme that ties the techniques together. Materials on MIT OpenCourseWare include lecture notes, problem sets, and exams from the MIT course taught within Electrical Engineering and Computer Science. The emphasis throughout is on understanding not just how algorithms work but why and when they succeed, moving from foundational supervised learning methods toward more advanced probabilistic models. No instructor video lectures are included, but the full set of written course materials lets a self-directed learner work through the curriculum independently, at no cost, with no certificate offered.