Georgia Institute of Technology
School of Electrical and Computer Engineering
ECE8833/BMED8813: Special Topics - Information processing models in neural systems
Syllabus |
Homework |
Calendar |
Course Materials |
Projects
Instructor:
Prof. Christopher J. Rozell
3105 U.A. Whitaker building
crozell -at- gatech -dot- edu
404.385.7671
Office hours: Tue 1:00-2:30 or by appointment
Course meeting: Tue/Thu 3:05-4:25 in Van Leer C340
Overview:
Neural systems allow biological organisms to navigate highly complex environments by efficiently processing and understanding their sensory information using computational architectures very different from digital computers. While data-driven modeling has been a valuable tool for describing the input/output relationships of various neural systems, the results do not always illuminate a system's underlying function that enables this high level of understanding. This course will examine cases where a computational principle describing an information processing strategy used in engineering (e.g., information theory, optimal detection/estimation, resource optimization) can account for observed properties in neural systems at levels ranging from the anatomy of single neurons to perception. The necessary neurophysiology and mathematics background material will be largely self-contained, making this course appropriate for students with a background in either the biosciences or engineering who are interested in learning how the tools of modern information processing can help us understand the function of neural systems.