Spring 2004 schedule: Tuesday and Thursday 11 - 12:30 in E51-085
Mathematical introduction to neural coding and dynamics. Convolution, correlation, linear systems, game theory, signal detection theory, probability theory, information theory, and reinforcement learning. Applications to neural coding, focusing on the visual system. Hodgkin-Huxley and related models of neural excitability, stochastic models of ion channels, cable theory, and models of synaptic transmission.
Prof. Sebastian Seung,
Office hours: TBA, E25-429
T.A.s Jennifer Wang (email@example.com) and Justin Werfel
Office hours: Tuesday, 4pm-5pm E25-425 (Jen) or by appointment
Optional lectures will be Mondays 7-8 PM in selected weeks, held in E25-401.
No optional lecture this week (3/8)
The central assumption of computational neuroscience is that the brain computes. What does that mean? Generally speaking, a computer is a dynamical system whose state variables encode information about the external world. In short, computation equals coding plus dynamics. Some neuroscientists study the way that information is encoded in neural activity and other dynamical variables of the brain. Others try to characterize how these dynamical variables evolve with time. The study of neural dynamics can be further subdivided into two separate strands. One tradition, exemplified by the work of Hodgkin and Huxley, focuses on the biophysics of single neurons. The other focuses on the dynamics of networks, concerning itself with phenomena that emerge from the interactions between neurons. Therefore computational neuroscience can be divided into three subspecialties: neural coding, biophysics of neurons, and neural networks.