CS 446

Principles and applications of machine learning. Main paradigms and techniques, including discriminative and generative methods, reinforcement learning: linear regression, logistic regression, support vector machines, deep nets, structured methods, dimensionality reduction, k-means, Gaussian mixtures, expectation maximization, Markov decision processes, and Q-learning. Application areas such as natural language and text understanding, speech recognition, computer vision, data mining, and adaptive computer systems, among others. Course Information: Same as ECE 449. 3 undergraduate hours. 3 or 4 graduate hours. Prerequisite: CS 225; One of MATH 225, MATH 257, MATH 415, MATH 416, ASRM 406 or BIOE 210; one of CS 361, STAT 361, ECE 313, MATH 362, MATH 461, MATH 463, STAT 400 or BIOE 310.

Machine Learning by Tom Mitchell, Intro. to Machine Learning by Ethem Alpaydin, Learning Kernel Classifiers by Ralf Herbrich, Kernel Methods for Pattern Analysis by Shawe-Taylor & Cristianini & C.M. Bishop, Neural Networks for Pattern Recognition by C.M. Bishop, An Intro. to Computational Learning Theory by Kearns & Vazirani, Learning Theory Learning with Kernels by Schlkopf & Smola, and Bayesian Reasoning and Machine Learning by David Barber.

Be able to articulate key concepts and principles in Machine learning (1), (2), (4), (5)
Be able to articulate and model problems given an understating of representational issues and abstraction in machine learning. (1), (2), (3), (5), (6)
Be able to explain and analyze models and results making use of theoretical principles and the limitations of generalization in machine learning. (1), (2), (3), (5), (6)
Make use of the algorithmic theory of machine learning in problem analysis and model selection. (1), (2), (3), (5), (6)
Understand and apply the maximum likelihood principle and explain algorithmic implications in modeling and problem-solving. (1), (2), (3), (5), (6)
Be able to use a variety of algorithmic techniques in machine learning. (1), (2), (3), (5), (6)
Be able to choose and use a variety of machine learning protocols in different situations. (1), (2), (3), (4), (5), (6)
Familiarity with deep networks and how to fit them to data. (1), (2), (3), (5), (6)

Introduction to Machine Learning
Learning Decision Trees
On Line Learning Algorithms
Features and Kernels
Computational Learning Theory
Boosting
Support Vector Machines
Multiclass Classification
Bayesian Learning and Inference
Semi-Supervised Learning and the EM algorithm
Learning Probability Distributions
Clustering
Deep learning

Revisions in last 6 years	Approximately when revision was done	Reason for revision	Data or documentation available?	Documentation provided?
Added SVMs and Kernels.	Several times, Fall 2009, 2010, 2011.	Emphsize growing understanding and importance of these topics.	Professional judgement	Course web site documents the updates
added Graphical Models and approcximate inference	Fall 2011	Emphsize growing understanding and importance of these topics	Professional judgement	Course web site documents the changes
took out Neural Networks and Rules	Fall 2011		Professional judgement	Course web site documents the changes

Selected Elective.

2/18/2019by Sanmi Koyejo