ECE 4950

Special Topics in Electrical and Computer Engineering

Course Description

Course information provided by the Courses of Study 2016-2017.

The course will be devoted to understanding, implementation, and applications of various machine learning primitives. We will cover several unsupervised, and supervised learning techniques. Supervised learning includes regression, support vector machines, decision trees, random forests, naïve bayes, boosting and bagging. Unsupervised learning includes clustering, k-means, k-NN, principal components analysis and other dimensionality reduction methods. We will also cover Mixture of Gaussians, the EM algorithm, and the Viterbi algorithm. We will give particular emphasis on applications in ECE, e.g., text data, hand-writing, music, image, and time series data, and categorical datasets such those in recommendation systems. The course will have a programming component, which will be administered in the form of assignments, or in-class-kaggle competitions.

When Offered Fall, spring.

Prerequisites/Corequisites Prerequisite: Linear Algebra (MATH 2940 or equivalent), Basic Probability and Statistics (ECE 3100, STSCI 3080, or equivalent).

Comments Seminar, special interest, or temporary course.

View Enrollment Information