CS 778 – Topics in Neural Networks and Machine Learning



Hopfield Networks

Required: Hopfield and Tank

Optional: Hopfield Chapter

Slides: Hopfield


Boltzmann Machines

Required: Original Boltzmann


Slides: Boltzmann


Speech Recognition


Speech Overview

HMM and Speech

Multcons Hopfield Extensions

Multcons Slides


Active Learning

Required: Improving Generalization with Active Learning

Optional: Survey Paper


Bayesian Model Averaging

Required: Bayesian Model Averaging and Overfitting

Optional: Statistics Field Survey Paper


Spiking Neural Networks

Required: Spiking Overview - Chapter 2 required, other chapters optional

Required: SpikeProp Paper


Ripper Rule Learning Algorithm

Required: Ripper

Optional: Mitchell, T., Machine Learning, Ch. 10.1-10.3


Feature Selection: Filters and Wrappers, etc.

Required: Wrappers and Filters

Optional: Overview of Feature Selection


Kalman Filters and Extensions

Required: Kalman Filter Overview - pp. 15-37 required, other optional

Optional: http://en.wikipedia.org/wiki/Kalman_filter

Optional: Training Neural Networks with the Extended Kalman Filter


Non-Linear Dimensionality Reduction

Required: Kernel PCA Algorithm

Optional: Longer Version Kernel PCA Algorithm Paper

Optional: Overview Paper of Manifold Learning (Non-Linear Dimensionality Reduction)


Bias/Variance Decomposition

Required: Bias Plus Variance Decomposition for Zero-One Loss Functions


Unsupervised Learning/Clustering

Required: Clustering Overview

Required: Incremental Concept Formation/Conceptual Clustering/Cobweb


Slides for study:


Rule Learning