Special Topics in Computer Science:
Machine Learning
CIS 595 – Spring 2002
Meeting days:
Monday, 7:25P - 9:55P, TL302
Instructor: Slobodan Vucetic, 323 Wachman Hall, phone: 204-5773, www.ist.temple.edu/~vucetic
Office Hours: Wednesday 2:00 pm - 3:00 pm, or by appointment
Objective: The goal of the field of machine learning is to build computer systems that learn from experience and that are capable to adapt to their environments. Learning techniques and methods developed by researchers in this field have been successfully applied to a variety of learning tasks in a broad range of areas, including, for example, text classification, gene discovery, financial forecasting, credit card fraud detection, collaborative filtering, design of adaptive web agents and others. Required
texts:
· T.M. Mitchell, Machine Learning, 1997.
· R.O. Duda, P.E. Hart, D.G. Stork, Pattern Classification, 2000.
Additional papers and handouts relevant to presented topics will be distributed as needed.
Topics:
· Concept learning. Version spaces. PAC learning. VC dimension.
· Regression. Loss function. Least-squares fit. Parameter estimation. Statistical view on the regression. Log likelihood measures. On-line learning techniques.
· Classification. Logistic regression. Class-conditional densities. Parameter estimation. Perceptron algorithm. On-line techniques. Multiple classes.
· Neural networks. Nonlinear decision boundaries. Backpropagation. Radial basis functions.
· Classification and regression trees CART. C4.5.
· Support vector machines. Classification. Max margin hyperplanes. Kernel functions.
· Ensemble methods. Mixture of experts. Bagging. Boosting.
· Unsurpervised learning. Clustering, k-means.
· Density estimation. Parametric methods. Mixture of Gaussians. Non-parametric. Parzen windows. Nearest neighbor.
· Dimensionality reduction. Feature extraction. Mutual information measure.
PCA. Clustering.
· Bayesian networks. Independence structure. Inference. Parameter learning. Structure learning.
· Learning with hidden variables and missing data. Expectation-maximization algorithm.
· Hidden Markov models. Forward, backward algorithm. Baum-Welch algorithm.
· Markov random fields. Independence structure. Inference. Learning.
· Reinforcement learning. Learning to act. Markov decision processes. Reinforcement learning with delayed rewards.
Grading:
A combination of homework assignments (40%), a midterm (20%), and an individual project (40%)