ECE6535

Course covers a set of mathematical and statistical methods that are fundamental for analyzing and modeling neural/cognitive data and neural signal and information processing, practiced through extensive computational exercises. Topics include linear algebra, least-squares regression, eigen-analysis and PCA, linear shift-invariant systems, convolution, Fourier transforms, Nyquist sampling, basics of probability and statistics, hypothesis testing, model comparison, bootstrapping, estimation and decision theory, signal detection theory, classification, linear discriminants, clustering, simple models of neural spike generation, analysis. Intended for students from quantitative backgrounds, i.e. engineering, math, statistics, computer science, physics, neuroscience, psychology.

Prerequisites: Instructor Permission.
Corequisites: ECE 5536 OR 6536

Working knowledge of calculus, algebra, and trigonometry is expected. Some experience with linear algebra, matrix computation, basic probability & statistics, and computer programming.

Fall