Statistical Machine Learning
CMPE Degree: This course is Not Applicable for the CMPE degree.
EE Degree: This course is Not Applicable for the EE degree.
Lab Hours: 0 supervised lab hours and 0 unsupervised lab hours.
Technical Interest Group(s) / Course Type(s): Digital Signal Processing
Catalog DescriptionAn introduction to the theory of statistical learning and practical machine learning algorithms with applications in signal processing and data analysis.
Textbook(s)The Elements of Statistical Learning
Student OutcomesIn the parentheses for each Student Outcome:
"P" for primary indicates the outcome is a major focus of the entire course.
“M” for moderate indicates the outcome is the focus of at least one component of the course, but not majority of course material.
“LN” for “little to none” indicates that the course does not contribute significantly to this outcome.
1. ( Not Applicable ) An ability to identify, formulate, and solve complex engineering problems by applying principles of engineering, science, and mathematics
2. ( Not Applicable ) An ability to apply engineering design to produce solutions that meet specified needs with consideration of public health, safety, and welfare, as well as global, cultural, social, environmental, and economic factors
3. ( Not Applicable ) An ability to communicate effectively with a range of audiences
4. ( Not Applicable ) An ability to recognize ethical and professional responsibilities in engineering situations and make informed judgments, which must consider the impact of engineering solutions in global, economic, environmental, and societal contexts
5. ( Not Applicable ) An ability to function effectively on a team whose members together provide leadership, create a collaborative and inclusive environment, establish goals, plan tasks, and meet objectives
6. ( Not Applicable ) An ability to develop and conduct appropriate experimentation, analyze and interpret data, and use engineering judgment to draw conclusions
7. ( Not Applicable ) An ability to acquire and apply new knowledge as needed, using appropriate learning strategies.
Strategic Performance Indicators (SPIs)
1. Supervised learning
a) The Bayes classifier and the likelihood ratio test
b) Nearest neighbor classification
c) Linear classifiers
i. plugin classifiers (LDA, logistic regression, Naïve Bayes)
ii. the perceptron learning algorithm
iii. maximum margin principle and separating hyperplanes
d) Linear regression
i. least-squares linear regression
ii. the LASSO
e) Theory of generalization
ii. concentration inequalities
iii. VC dimension and generalization bounds
iv. the bias-variance tradeoff
f) Nonlinear classifiers
i. nonlinear feature maps
ii. the kernel trick
iv. multi-layer neural networks
g) Nonlinear methods in regression
h) Error estimation and validation
2. Unsupervised learning
a) Linear dimensionality reduction and principal component analysis
b) Mutltidimensional scaling
ii. GMMs and the EM algorithm
iii. spectral clustering
d) Density estimation
e) Feature selection
f) Nonlinear dimensionality reduction (manifold learning)
3. Other topics (as time permits)
a) Matrix factorizations
b) Graphical models
c) Ensemble methods (boosting, random forests)
d) - Deep learning