Code for Neural Networks and Reinforcement Learning (Post-Graduate)  
This site contains code for use in neural network research. The code illustrates how to train neural networks using MATLAB and C. There is code that illustrates how to initialize and train a recurrent neural network using Williams and Zipser's Real-Time Recurrent Learning algorithm. In addition, there are files that can be used for training feedforward neural networks with a single hidden layer using error backpropagation with early stopping using MATLAB and Octave.

Professor Name: Charles W. Anderson
Department: Computer Science
University: Colorado State University

Submitted: Aug 22, 2007
Introduction to Computational Neuroscience  
Description: This course focuses on mathematical concepts and techniques used in computational neuroscience. It is designed to provide students with necessary mathematical background for formulating, simulating, and analyzing models of individual neurons and neural networks. The course will serve as an introduction to the theory of nonlinear differential equations and applied dynamical systems in the context of neuronal modeling. The topics to be covered include a review of basic facts about the electrophysiology of neural cells, analysis of the conductance based models, neural excitability, bursting, models for synaptically coupled cells, and compartmental models, as well as a number of mathematical techniques such as phase plane analysis, fast-slow decomposition, and elements of the bifurcation theory. The students will learn basic models of excitable membranes such as Hodgkin-Huxley, Morris-Lecar, and FitzHugh-Nagumo models.
Course material created by Professor Georgi Medvedev.
Target audience: Graduate
Institution: Drexel University
Materials available: Problem sets or projects, Course outline or syllabus, Textbook recommendations, Downloadable code or data files
Products: MATLAB

Submitted: Jul 30, 2008
Hodglom-Huxley-like Model Neuron  
Description: This file provides an interactive MATLAB-based demo simulation in which the electrophysiological behavior of a biological neuron is presented and documented.
Target audience: Advanced undergraduate (3rd or 4th year)
Academic institution: N/A
Materials available: Presentations, Downloadable code/data files
Products: MATLAB

Submitted: Jul 08, 2008

  Privacy - Trademarks - Feedback - Terms of Use Copyright The MathWorks, Inc.