Course Content |
|
- Supervised Learning Algorithms:
- Logistic Regression,
- Neural Networks,
- Decision Trees,
- Nearest Neighbour,
- Support Vector Machines, and
- Naive Bayes.
- ML and MAP estimates.
- Bayes’ Optimal Classifier.
- Introduction to Graphical Models.
- Generative Vs. Discriminative Models.
- Unsupervised Learning Algorithms:
- K-Means clustering,
- Expectation Maximization, and
- Gaussian Mixture Models.
- PCA and Feature Selection.
- PAC Learnability.
- Reinforcement Learning.
- Some application areas of machine learning e.g. Natural Language
Processing, Computer Vision, applications on the web.
- Introduction to advanced topics such as Statistical Relational Learning.
|
|
Textbooks |
|
- For the most part we will follow "Learning From Data", by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin
- For the topics not covered above we will follow "Machine Learning", by Tom Mitchell
Other useful reference books:
- Hastie, Tibshirani, and Friedman: The Elements of Statistical Learning.
- Christopher Bishop: Pattern Recognition and Machine Learning.
- Kevin Murphy: Machine Learning: a Probabilistic Perspective
- Shai Shalev-Shwartz and Shai Ben-David: Understanding Machine Learning: From Theory to Algorithms.
- David Barber: Bayesian Reasoning and Machine Learning
- Richard S. Sutton and Andrew G. Barto: Reinforcement Learning: An Introduction
|
|
Course Policies |
|
Prerequisites
- As per courses of study
- The course is expected to be assignment heavy, and a student may be required to install and work with multiple libraries, at times through public web services. Proficiency in programming and working with multiple external libraries would be desired.
|
|
(Tentative) Marks Distribution
|
- Minor 1: 10
- Minor 2: 10
- Major: 20
- Assignments(3-4)*: 60
*All assignments may not be of similar marks. We will announce percentage marks of an assignment at the time of releasing it.
|
|
Other Policies
|
- Auditing the course is highly discouraged. You should choose to do a sit through instead. We will keep a minimum of B grade in the course for Audit pass, along with a requirement to score 30% in each of the assignment.
- You can sit through a course with explicit permission from me (over email). You can attend lectures but can not submit assignments or sit in the exam.
- Any plagiarism detected in any of the assignments will lead to zero in the assignment, and a grade reduction (e.g. B to B-) on top of it.
- Other institute rules such as failing for skipping major exam will also apply
- No deadline extension in any assignment submission
|