Introduction to Machine Learning

Instructor	Hongjie Chen, jeffchan@vt.edu, 540-818-1850
Class Hours	Monday, Tuesday, Thursday 3:30 pm - 5:35 pm at Class Zoom
Teaching Assistant	Aman Ahuja, aahuja@vt.edu
Office Hours	Hongjie Chen: Tuesday 10:00 am - noon at Hongjie's Personal Zoom Aman Ahuja: Wednesday and Friday 3:00 pm-4:30 pm at Aman's Personal Zoom
Course Websites	Piazza, Canvas CS4824, Canvas ECE4424

Description

Machine Learning is everywhere in our daily life: from face recognition lock and video recommendation to language translation and speech agent interaction, we can be surprised by how many products are powered by machine learning techniques. In fact, there are much more domains (e.g., bioinformatics, robotics, system security) that leverage machine learning than the listed examples.
This course will cover fundamental topics in Machine Learning by introducing key problems, intuitions to solutions, mathematical foundations, and realistic applications.

Topics

Basics of Statistical Learning

Loss functions, MLE, MAP, Bayesian estimation, bias-variance tradeoff, overfitting, regularization, cross-validation

Supervised Learning

Decision Trees, Naïve Bayes, Logistic Regression, Regression, Kernels and Kernel Regression, Support Vector Machines, Neural Networks

Unsupervised Learning

Expectation Maximization, Gaussian Mixture Clustering, K-Mean Clustering

Graphical Models

Bayesian Networks, Hidden Markov Models

Deep Learning

Convolutional Neural Networks, Recurrent Neural Networks, Attention and Transformer Networks, Autoencoders, Variational Autoencoders, Generative Adversarial Networks

Reinforcement Learning

Markov Decision Process, Value Iteration, Policy Iteration, Q-Learning

Machine Learning Applications

Time-series, Graph Machine Learning

Prerequisites

Ability to deal with abstract mathematical concepts.
Probability and Statistics

Basic concepts of probability including random variables, expectation, conditional distribution, Bayes rule, chain rule, likelihood, prior probability, densities, marginalization, moments, etc.

Calculus and Linear Algebra

Matrix multiplication, eigenvalues, positive semi-definiteness, multivariate derivatives.

Algorithms

Basic data structures, complexity analysis

Programming

This is a demanding class in terms of programming skills. Homework assignments will involve a mix of Python and libraries.

Please speak with the instructor if you have a concern on your background.

Class Goal

A student who successfully completes this class should

understand fundamental machine learning concepts;
be able to implement standard machine learning methods from scratch;
be able to recognize tasks that could be solved by machine learning models;
be able to formulate applicable tasks and connect them with machine learning methods;
and have the background knowledge to be able to understand new machine learning methods not covered in the course.

Grading

Homework (Five coding assignments): 95%
Class participation: 5%
No written exam

Based on the grading breakdown above, each student's final grade for the course will be determined by the final percentage of points earned. The grade ranges are as follows:

A:	93.3%–100%,	A-:	90.0%–93.3%,	B+:	86.6%–90.0%,	B:	83.3%–86.6%,	B-:	80.0%–83.3%,	C+:	76.6%–80.0%,
C:	73.3%–76.6%,	C-:	70.0%–73.3%,	D+:	66.6%–70.0%,	D:	63.3%–66.6%,	D-:	60.0%–63.3%,	F:	00.0%–60.0%.

Textbooks are not required.
Optional reference books (freely available online):

Machine Learning: a Probabilistic Perspective, Kevin Murphy, MIT Press, 2012
Pattern Recognition and Machine Learning, Christopher Bishop, Springer, 2006
The Elements of Statistical Learning, Trevor Hastie, Robert Tibshirani, and Jerome Friedman, Springer, 2009
Deep Learning, Ian Goodfellow, Yoshua Bengio, and Aaron Courville, MIT Press, 2016
Reinforcement Learning: An Introduction, Richard S. Sutton and Andrew G. Barto, MIT Press, 2018

Schedule

Note: This schedule is tentative and subject to change. All due dates are until 11:59 PM EDT.

Week / Day	Date	Topic	Readings and Notes
1^st / Monday	May 23^rd	Class Logistics, Function Approximation, Decision Tree	HW 1 Out, C.2.6
1^st / Tuesday	May 24^th	Model Selection, Probability Estimation	C.2.9, C.7
1^st / Thursday	May 26^th	MLE and MAP, (Gaussian) Naïve Bayes	A.3
2^nd / Monday	May 30^th	......	No Class in Observance of Memorial Day
2^nd / Tuesday	May 31^st	Logistic Regression	HW 1 Due, HW 2 Out, A.8
2^nd / Thursday	June 2^nd	Generative v.s. Discriminative Classifiers	B.4
3^rd / Monday	June 6^th	Regression, Perceptron	A.7, A.8.5.4
3^rd / Tuesday	June 7^th	Kernel, Kernelized Perceptron	HW 2 Due, HW 3 Out, A.14, B.6, C.6
3^rd / Thursday	June 9^th	Neural Networks	A.16.5, A.14.5, B.5, B.7.1.4, C.12.3, C.11
4^th / Monday	June 13^th	Support Vector Machine (SVM), Graphical Models	A.10, B.8, C.17
4^th / Tuesday	June 14^th	Expectation Maximization (EM)	A.11, B.9, C.8.5
4^th / Thursday	June 16^th	Gaussian Mixture Clustering, K-Means	HW 3 Due, HW 4 Out, A.11.2.1, B.9, C.8.5
5^th / Monday	June 20^th	......	No Class in Observance of Juneteenth
5^th / Tuesday	June 21^st	Deep Neural Networks (DNN), Convolutional Neural Networks (CNN)	D.7, D.9
5^th / Thursday	June 23^rd	Recurrent Neural Networks (RNN), Autoencoders	HW 4 Due, HW 5 Out, D.10.2, D.14
6^th / Monday	June 27^th	Generative Adversarial Networks (GAN)	D.7.13
6^th / Tuesday	June 28^th	Reinforcement Learning	E.6
6^th / Thursday	June 30^th	Summary	HW 5 Due

Assignments

Homeworks (19% x 5)
Students are expected to work individually on 5 homework assignments throughout the semester. Assignments will involve coding implementation and analysis, covering various topics that complement and supplement the lecture topics. Assignments will involve a mix of Python and libraries to be submitted electronically via Canvas. Each assignment is graded in a scale of 100 points and counts 19% of the final grade.

Class participation (5%)
Students are strongly encouraged to attend all the lectures (exceptions are allowed due to medical reasons or emergencies) and expected to engage in the discussion during the lectures and participate in Q&A. Please inform the course staff via email if you cannot make it to the class. Students are also expected to be actively engaged in class-related discussion on Piazza so that other students may benefit from your questions and our answers. Your class participation grade will depend on your overall engagement in the classroom and in Piazza as well as your intellectual contribution. Participation counts 5% of the final grade.

Note: Students' first point of contact is Piazza (so that other students may benefit from your questions and our answers). If you have a personal matter, create a private piazza post or send an email to the instructor.

Policies

Late Submission Policy
Due to the short duration nature of the semester, submissions should be submitted before each assignment due to be graded. A one-time late submission ticket can be used to submit an assignment late but no later than the following midnight of the original due, as a no penalty late submission. When the ticket is used, a late submission that is no later than the following midnight of the original due will be accepted with 50% point off. One exception is that no late submission is allowed for homework assignment 5 due to the fact it is close to the end of the semester. All late submission must be submitted by emailing the instructor (Hongjie Chen) with an appropriate reasoning and attached files. Any other late submission will not be graded and will be treated as missing. Missing submission will be given zero point. No penalties will be given for medical reasons (show doctor's note) or emergencies. In these cases, students should notify the instructor as timely as possible to allow due time extension.

Regrading requests
Requests for regrading due to grading errors must be submitted to the TA and instructor via email within three days of the release of grades.

Academic integrity
The Undergraduate Honor Code pledge that each member of the university community agrees to abide by states: "As a Hokie, I will conduct myself with honor and integrity at all times. I will not lie, cheat, or steal, nor will I accept the actions of those who do."
Students enrolled in this course are responsible for abiding by the Honor Code. A student who has doubts about how the Honor Code applies to any assignment is responsible for obtaining specific guidance from the course instructor before submitting the assignment for evaluation. Ignorance of the rules does not exclude any member of the University community from the requirements and expectations of the Honor Code. For additional information about the Honor Code, please visit: https://www.honorsystem.vt.edu/
This course will have a zero-tolerance philosophy regarding plagiarism or other forms of cheating. Your homework assignments must be your own work, and any external source of code, ideas, or language must be cited to give credit to the original source. I will not hesitate to report incidents of academic dishonesty to the Office of the Undergraduate Honor System.

Principles of Community
Because the course will include in-class discussions, we will adhere to Virginia Tech Principles of Community.

Accessibility
If any student needs special accommodations because of any disabilities, please contact the instructor during the first week of classes. Such students are encouraged to work with The Office of Services for Students with Disabilities to help coordinate accessibility arrangements.

Disclaimer: Many of class slides are derived from those of Tom Mitchell, Pascal Poupart, Pieter Abbeel, Eric Eaton, Carlos Guestrin, William Cohen, and Andrew Moore. Much of the class structure is moderately inspired by that of Dr. Debswapna Bhattacharya's.