Course Description

Machine learning has been used in a wide variety of fields and has the great potential to solve the problems that we care about, such as climate change, misinformation, and pandemic predictions. Due to its importance, it is well agreed that every student should understand the impact of machine learning on her own field, and everyone should have an opportunity to learn it. This course is dedicated to non-computing major students. The primary focus is on giving an introduction to machine learning techniques applying to various problems. The primary goal of this course is to help students with non-computing backgrounds to build a mindset of how machine learning can help solve real-world problems. Upon successful completion of this course, students will be able to:

Understand the basic methodology in machine learning.
Implement the typical pipeline of using machine learning to solve problems.
Formulate real-world applications as machine learning problems and build end-to-end solutions using open-source libraries such as Scikit-Learn, Keras, and TensorFlow.

This course is called Machine Learning for All: From examples to algorithms, to emphasize two unique design features: (1) machine learning should be accessible to everyone and (2) teaching should start from examples instead of theory.

Background Requirements

The population of target students is undergraduates who have finished their first-year course requirements in UVa. The prerequisites of this course will be the Introduction to Programming (CS 1110) or the Software Development Methods (CS 2110). Consult the instructor to determine if you are sufficiently prepared or have any questions.

Instructor

Name: N. Rich Nguyen, Ph.D. | Assistant Professor

Email: nn4pj@virginia.edu

Office: Rice 202 or on Zoom

Office Hours: Tue & Thu 11a-12p, Wed 2-3p or by appointment

You may call me Professor Nguyen \: win\.
If you email me, please put [CS2501 ML4All] somewhere in the subject. Failure to do so makes it much harder to keep up with your email and reduces the chance of a timely reply.
Please don’t hesitate to contact me if you have any problems, concerns, questions, or issues regarding the course, material, or anything else in the class. I want to help!

Teaching Assistant

Medhini Rachamallu, 3rd-year Electrical/Computer Engineering student, mr3dn@virginia.edu

Piazza Discussion Board

This term we will be using Piazza for class discussion. The system is highly catered to getting you help fast and efficiently from classmates, the TA, and myself. Rather than emailing questions to the teaching staff, I encourage you to post your questions on Piazza. If you have any problems or feedback for the developers, email team@piazza.com.

Find our class signup link at: https://piazza.com/virginia/fall2021/cs2501ml4allf21

Textbooks

Aurelien Geron. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition. O'Reilly, September 2019 (Order here or get access to O’Reilly)

(Optional) Mark Fenner. Machine Learning with Python for Everyone, 1st Edition. Addison-Wesley Professional. August 2019 (Order here or get access for free via O’Reilly).
(Optional) Pedro Domingos. The Master Algorithm (Basics Book: 2015).

Course Topics

As shown in the table below, a tentative course plan covers six essential machine learning topics. Each lecture will focus on (1) introducing a machine learning problem from some examples, (2) building the connection between the machine learning problem and the algorithms, (3) the basic idea of each algorithm and the common implementation pitfalls in practice.

Topics	Examples	Algorithms
1. Classification	Misinformation identification, image classification	Logistic regression, naive Bayes classifiers, neural network classifiers
2. Regression	COVID-19 forecasts, stock market predictions	Linear regression, neural network regressors
3. Clustering and Density Estimation	Medical image segmentation, place-based predictive policing	k-means clustering, Gaussian mixture models, hierarchical clustering
4. Dimensionality Reduction	Financial document retrieval, social network visualization	Principal component analysis, neural encoder-decoder models
5. Sequential Modeling	Automatic text generation, traffic prediction	Markov models, recurrent neural networks
6. Ethics in Machine Learning	Facial recognition auditing, bias in student loan service	Local interpretable model-agnostic explanations (LIME), debiasing algorithms

In addition to the lectures, we also propose to have an online lab time at least once in two weeks. The primary purpose of this lab time is to give students some hands-on experience on implementing specific algorithms and also allow instructors (including TAs) to help students closely. Designing the exercise questions is also one of the specific tasks in this project. To help students focus more on implementation skills, we plan to provide a guideline with different levels of detail. In general, the guideline will have specific instructions about what they should do in each step.

Grading Scheme

A total of 1000 points to be given. The grading procedure is as follows: F: 0-599; D-: 600-629; D: 630-669; D+: 670-699; C- : 700-729; C : 730-769; C+ 770-799; B- : 800-829; B: 830-869; B+ 870-899; A- : 900-929; A: 930-969; A+ : 970-1000. The breakdown of the grade is as follows:

3 Homeworks: 3x150 pts (45%)
2 Surveys: 2x150 pts (30%)
1 Group Project: 250 pts (25%)

Overall, there are three forms of evaluation: surveys, homework, and a group project. As we value more about students’ practical skills in this course, we do not plan to give examinations.

Homework Assignments (45%)

On the pipeline of using machine learning in practice: as part of the homework, we will design some questions based on real-world applications. There will be 3 homework assignments. The students will be asked to formulate each question using machine learning terminology. For example, students should answer that predicting the sentiment polarity of a product review can be formulated as a classification problem.

If the results you see when running your program do not match the TA reports for your program, you should promptly talk to the TA. However, in all but the rarest of cases, you should be prepared to understand that the definitive test of your code is that made by the TA, and what you are seeking in talking to the TA is an understanding of how to avoid losing points over similar discrepancies in the future. To put this another way, you should contact the TA, but if you understand why there is a difference, you should not assume that your assignment will be re-graded. In general, it will not.

Late Policy: Deadlines are strictly enforced, but I understand that unexpected situations may arise which prevent you from submitting your assignment on time. You will receive a late penalty for each day in which your assignment is overdue at the rate of -10% per day. If you fail to turn in your overdue assignment after 3 days, you get zero points for that assignment.

Surveys (30%)

On the understanding of basic methodology: we will distribute a survey to students and ask them how they think about machine learning and how machine learning can help them solve problems. The survey will be distributed at least twice during the semester: one is at the beginning of the semester, and the other is at the end of the semester. The response difference will help us identify whether students understand the basic methodology of machine learning.

Course Project (25%)

On the skills of solving real-world problems: at the beginning of the course, students will be asked to bring a problem of their interests, which will be used as the problem for their final project. Along with the lectures, they will be asked to identify some machine learning methods that they learn from this course to help solve this problem. At the end of this course, they will use this problem as their final project, and the solution will be presented in a workshop particularly organized for this class. The course project to be undertaken is called Machine Learning For Virginia (ML4VA). Details of the groups are:

The required group size is three. Individuals may be able to work by themselves with good reason. Expectations will not be adjusted due to smaller group sizes. Groups of more than three are strongly discouraged but can be allowed under extreme circumstances.
In general, all group members will receive the same grade for graded assignments. However, group members will evaluate their peers and any student who appears not to be contributing may be penalized.
Each group will be responsible for assigning tasks to its group members.

You are expected to work as a member of your group in this course and cooperate with your colleagues. There will be multiple working sessions and checkpoints to help your team stay on top of the project. Cooperation means attending group meetings, completing your assignments properly, and letting your group know if you will be out of town, responding to emails from your group, and so on. If there is a lack of cooperation by any group member, it must be brought to the instructor’s attention as soon as it happens. If the lack of collaboration is serious, the offending group member’s semester grade will be lowered.

Extra Credits (up to 10%)

There will be multiple opportunities including code-a-thons and hackathons throughout the semester to receive extra credits (up to 10% of the total grade). The extra credits can only be given at the instructor’s discretion. Please be sure to provide sufficient documentation and report in order to receive the credits. Failure to complete or provide documentation to the satisfaction of the instructor does not result in any extra credits.

Academic Integrity

The School of Engineering and Applied Science relies upon and cherishes its community of trust. We firmly endorse, uphold, and embrace the University’s Honor principle that students will not lie, cheat, or steal, nor shall they tolerate those who do. We recognize that even one honor infraction can destroy an exemplary reputation that has taken years to build. Acting in a manner consistent with the principles of honor will benefit every member of the community both while enrolled in the Engineering School and in the future.

Students are expected to be familiar with the university honor code, including the section on academic fraud (http://www.virginia.edu/honor/what-is-academic-fraud-2/). Unless otherwise noted, exams and individual assignments will be considered pledged that you have neither given nor received help. (Among other things, this means that you are not allowed to describe problems on an exam, assignment, or project to a student who has not taken it yet. You are not allowed to show exam papers to another student or view another student’s exam papers while working on an exam.) Sending, receiving, or otherwise copying or describing the contents of electronic files that are part of course assignments are not allowed collaborations (except for those explicitly allowed in assignment instructions). Here are some general guidelines:

In general, we expect that you will be using code, examples, and ideas from many different websites and resources for your projects. This is allowed within reason. Wholesale copying of an entire project is definitely not allowed. Using code to round out a feature is allowed. If you ever have a question about what is or is not appropriate, ask first!
In ALL cases, you need to cite all sources at the top of the file where the code or algorithm was used AND you should note all sources in your documentation.
Failure to properly attribute your sources will result in a 0 for the project at a minimum.

Attendance in lectures is vital to learning the material and making a good grade in this class.
Students can miss one required attendance day with no penalty. The second miss incurs a small penalty, with the penalty size increasing with each missed required class.

Professionalism

In this course, there will be a focus on working well together and the learning process. A large portion of that process involves interpersonal skills and conflict management. Students and staff are all expected to treat each other with respect.

The number one problem with professionalism in class is the overuse of laptops and mobile devices. Taking notes on a laptop and following along with the slides is welcome and encouraged. Doing work for other classes, watching videos, chatting, or anything else that distracts from your ability to learn and follow along (or anyone around you), will result in a professionalism penalty.

If you need to do work for another class during our class meeting, do so somewhere else and listen to the podcast, rather than disrupt the class itself. Students can and will be penalized for unprofessional behavior.

On Your Well Being

COVID-19 Guideline: This course will follow closely the University’s guidelines on Coronavirus. You can find the most updated policy here: https://coronavirus.virginia.edu/updates/important-information-about-fall-2021

The Engineering School proudly serves as a safe space for its students and aims to promote their well-being. If you are feeling overwhelmed, stressed, or isolated, there are many individuals here who are ready and wanting to help. In addition to the course instructors, you can seek help through the Engineering Undergraduate office (Thornton A122), or Alex Hall (aec5d, 924-7601) who is the assistant dean of students for the Engineering school. Alternatively, there are also other University of Virginia resources available. The Student Health Center offers Counseling and Psychological Services (CAPS) for its students. Call 434-243-5150 (or 434-972-7004 for after-hours and weekend crisis assistance) to get started and schedule an appointment. If you prefer to speak anonymously and confidentially over the phone, call Madison House's HELPLine at any hour of any day: 434-295-8255.

If you or someone you know is struggling with gender, sexual, or domestic violence, there are many communities and University of Virginia resources available. The Office of the Dean of Students, Sexual Assault Resource Agency (SARA), Shelter for Help in Emergency (SHE), and UVA Women's Center are ready and eager to help. Contact the Director of Sexual and Domestic Violence Services at 434-982-2774.
If you have been identified as an SDAC/LNEC student, please let the Center know you are taking this class. If you suspect you should be an SDAC/LNEC student, please schedule an appointment with them for an evaluation. We happily and discreetly provide the recommended accommodations for those students identified by the LNEC. Please contact us one week before an exam so we can make accommodations. Website: http://www.virginia.edu/studenthealth/sdac/sdac.html
If you have other special circumstances (athletics, other university-related activities, etc.) please contact your instructor and/or Head TA as soon as you know these may affect you.

About This Syllabus

This syllabus is considered a reference document that can and will be adjusted through the course of the semester to address changing needs. It is up to the student to monitor this page for any changes as this syllabus can be changed without notification. The final authority on any decision in this course rests with the course instructor, not with this document.

CS 2501 Syllabus - Page