ECE 63200 - Machine Learning and Data Mining
Course Details
Credits: 3
Areas of Specialization:
- Education
Counts as:
Normally Offered:
Fall - odd years
Catalog Description:
Machine learning is concerned with computer programs that automatically improve their performance through experience. Knowledge discovery in databases is concerned with extracting useful patterns or deviations from data using "data mining" methods. This course introduces students to the primary approaches to machine learning and data mining from a variety of fields, including inductive inference of decision trees, neural network learning, statistical learning methods, reinforcement learning, clustering, and discovery. In addition this course will introduce theoretical concepts such as inductive bias and the PAC (Probably Approximately Correct) learning framework.
Required Text(s):
None.
Recommended Text(s):
None.
Lecture Outline:
1 | Introduction: What is machine learning? Concept formation |
2 | Decision trees: test selection, pruning, MDLP, Increment versus Bach |
3 | Instance-based learning; logically weighted regression |
4 | Neural networks: Perceptrons and gradient descent, backpropagation |
5 | Bayesian approaches: Basics, EM, hidden Markov models |
6 | Knowledge discovery in databases |
7 | Empirical evaluation of learning systems |
8 | Boosting, feature selection |
9 | Computational learning theory |
10 | Scientific discovery; deviation detection |
11 | Clustering |
12 | Reinforcement learning; Q-learning; TD-learning |
13 | Learning from time series |
14 | In Class presentation of course projects |
15 | In class presentation of course projects, exams |
Assessment Method:
none