teaching graduate machine learning (UK CS 685)
01 January 2012
Course Summary
Machine learning is the branch of artificial intelligence that focuses
on algorithms that enable computer systems to learn from experience
and adapt to their environments. Machine learning algorithms and
techniques have been successfully applied to a wide range of domains
including text, audio, imagery, video, web click streams, social
networks, operating system scheduling, credit card fraud detection,
and targeted advertising placement.
This course is a self-contained introduction to the field of machine
learning. Students taking this course will learn both theoretical and
practical issues in the design of learning systems. Students will
derive and prove foundational theories and they will implement
classical and modern learning algorithms. In addition, students will
have significant freedom in choosing a final project, potentially
using problems from their personal research interests.
Course topics are drawn from the following list:
- Core Topics: Regression, Classification, Decision Trees, Neural
Networks, Support Vector Machines, Generative and Discriminative
Learning, Kernel Methods, Boosting, Clustering, Learning Theory,
Bias/Variance Trade-off, Expectation-Maximization
- Extended Topics: PCA, ICA, ISOMAP, SDE, Semi-Supervised
Learning, Distance Metric Learning, Reinforcement Learning, Online
Learning