East Carolina University
Department of Computer Science

CSCI 4120
Machine Learning
Standard Syllabus

Catalog entry

P: CSCI 2540; MATH 2228 or MATH 2283. Machine learning and statistical pattern recognition algorithms and their application to data analytics, bioinformatics, speech recognition, natural language processing, robotic control, autonomous navigation, and text and web data processing.

Course summary

Do you wonder about how IBM Watson, a question-answering system that responds to questions posed in natural language, won Jeopardy game championship in 2011?

Can you automatically generate textual descriptions that reflect the content of digital images?

How do you automatically colorize black and white movies?

How do you achieve real-time transnational of speech given in one language to another?

How do you discover patterns hidden in large data and use them to improve sales?

How do you make predictions based on historical data?

All of the above and more is possible using machine learning. In this course, you will learn theory, algorithms, and tools that enable you to solve problems like the above. Why are you waiting then?

Course topics

• Basic concepts of machine learning and example applications
• Learning theory - bias/variance trade-offs, Union and Chernoff/Hoeffding bounds
• Supervised learning: logistic regression, Perceptron, naive Bayes, generative learning algorithms, Gaussian discriminant analysis, support vector machines, model and feature selection, and ensemble methods
• Unsupervised learning: clustering, K-means, EM, mixture of Gaussians, factor analysis, principal components analysis, independent components analysis
• Reinforcement learning and control : MDPs, value and policy iterations, linear quadratic regulation, Q-learning, value function approximation
• Deep learning

Student learning outcomes

• Demonstrate a broad understanding of various machine learning and statistical pattern recognition algorithms and their application to diverse practical problems.
• Apply supervised learning algorithms (parametric/non-parametric algorithms, support vector machines, kernels, and neural networks) to solve practical problems.
• Apply unsupervised learning algorithms (clustering, dimensionality reduction, recommender systems, and deep learning) to solve practical problems.
• Apply best practices in machine learning (bias/variance theory) to solve diverse problems in domains ranging from computer vision to speech recognition, information retrieval, and natural language processing.

Textbook

Other required material

• David Barber. Bayesian Reasoning and Machine Learning. New York, NY: Cambridge University Press, 2012. isbn: 978-0521518147.
• Trevor Hastie, Robert Tibshirani, and Jerome Friedman. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Second. New York, NY: Springer, 2013. ISBN: 978-0387848570.
• Yaser S. Abu-Mostafa, M. Magdon-Ismail, and H-S. Lin. Learning From Data. Berlin, Germany: AMLBook, 2012. ISBN: 978-1600490064.
• Kevin Murphy. Machine Learning: A Probabilistic Perspective. Adaptive Computation and Machine Learning. Cambridge University Press: The MIT Press, 2012. ISBN: 978-0262018029.
• Christopher Bishop. Pattern Recognition and Machine Learning. New York, NY: Springer, 2007. ISBN: 978-0387310732.

Grading

Course grade is based on four components:

Grade meanings