EE 676 Neural Information Processing Systems

Basic neural computational models: Basic concepts of neural networks, inference and learning, classification models, association models, optimization models, self-organization models. Learning: supervised and unsupervised: AI learning, neural network learning-backpropagation, Generalization methods, radial basis function networks, reinforcement learning, genetic algorithms. Learning: incremental symbolic methods- COBWEB, decision tree approaches, neural network approaches, probabilistic neural networks, cascade correlation learning. Complex domains: hierarchical models- neocognitron, complex networks, modular neural networks, differentiation models-Kohonen's self-organizing nets. Learning spatiotemporal patterns. Neural networks and expert systems: expert systems, hybrid expert systems, fuzzy logic and neural networks. Rule generation from neural networks. Learning grammars: formal grammars, the neural network approach.

Credits

3

Prerequisite

EE 605