CS 4641\7641 A: Machine Learning (Summer 2021)

Course Information

• Lecture: Mondays and Wednesdays, 9:30am-11:40am
• Q/A for Lecture: Mondays, 5:00pm-6:00pm - Offline Questions
• Q/A for HWs: Wednesdays, 2:00pm-3:00pm - Use Edstem for offline questions
• Location: BlueJeans
• Edstem: https://edstem.org/us/courses/5944/

Course Overview

This course introduces techniques in machine learning with an emphasis on algorithms and their applications to real-world data. We will investigate the following question: how to computationally extract useful knowledge from data for decision making and task support? We will focus on machine learning methods, which are organized into three parts:

1. Basic math for data science and machine learning

   • Linear algebra
   • Probability and statistics
   • Information theory
   • Optimization

2. Unsupervised machine learning for data exploration

   • Clustering analysis
   • Dimensionality reduction
   • Kernel density estimation

3. Supervised learning for predictive data analysis

   • Tree-based models
   • Support vector machines
   • Linear classification and regression
   • Neural networks

Prerequisites for this course include (1) basic knowledge of probability, statistics, and linear algebra; (2) Basic programming experience in Python.

In addition to the technical content, this class includes the following learning objectives:

• Structuring a task into a machine learning work flow
• Collaborating effectively on team projects in a remote environment
• Conducting peer evaluation in a constructive format
• Communicating technical content in a concise and effective manner

Schedule

For all dates used in this course, their times are 23:59 Anywhere on Earth (11:59 pm AoE). For example, a due date of "January 8" is the same as "January 8, 23:59pm AoE". Convert the times to your local times using a Time Zone Converter.

Date	Topic	Assignment	Due	Quizzes	Readings
May 17, 2021	*Course Overview; Refer to Lecture1 video on Canvas Refer to Lecture1-QA video on Canvas	Edstem Signup			GT Honor Code
May 19, 2021	*Foundations: Linear Algebra; *Class Notes; Refer to Lecture2 video on Canvas				Correlation vs Covariance Linear Algebra Review by Zico Kolter
May 21, 2021				Q0 (warmup) Q1(Linear Algebra)
May 24, 2021	*Foundations: Probability and Statistics; *Class Notes; *QA Notes; Refer to Lecture3 video on Canvas Refer to Lecture3-QA video on Canvas	A1 Out			Probability Theory Review by Andrew Moore; The Differences Between Data, Information and Knowledge;
May 26, 2021	*Foundations: Information Theory; *Foundations: Optimization; *Information Theory Class Notes; Refer to Lecture4 video on Canvas		Team Creation Due		The Differences Between Data, Information and Knowledge; More about Cross Entropy and KLD; Cross Entropy as loss function; KKT for inequality constrained optimization; Why Cross Entropy over MSE for Classification; Gradient Descent short video
May 28, 2021				Q2(probability) Q3(info-theory)
May 31, 2021	No Class - Memorial Day
June 2, 2021	*Foundations: Optimization; *Data Analysis Toolbox; *Broadcasting Toolbox; Refer to Lecture5 video on Canvas				Spring 2021 Projects; Heilmeier catechism; Visual Information Theory by Chris Olah; GitHub Pages; YAML Configuration; NumPy Tutorial; Matplotlib Tutorial; Project Examples; seaborn: statistical data visualization; Overleaf for GT students; NumPy Tutorial; Matplotlib Tutorial;
June 4, 2021				Q4(Optimization) Q5(Toolbox)
June 7, 2021	*Clustering Analysis and K-Means; *K-Means Class Notes; *Hierarchical Clustering; *Hierarchical Class Notes; NO QA Notes; Refer to Lecture6 video on Canvas Refer to Lecture6-QA video on Canvas				Curse of dimensionality (Euclidean space example); Jupyter Notbook (Kmeans and DBSCAN); Understanding the concept of Hierarchical clustering Technique; Dendrogram Visualization ;
June 9, 2021	*Gaussian Mixture Model; *Class Notes; Refer to Lecture7 video on Canvas	A2 Out	A1 Due		GitHub Student Application;
June 11, 2021				Q6(kmeans) Q7(hierarchical)
June 14, 2021	*Gaussian Mixture Model; *Density-Based Clustering; *Class Notes (DBSCAN); *QA Notes [No notes]; Refer to Lecture8 video on Canvas Refer to Lecture8-QA video on Canvas [No video]		Proposal: 3 minutes recorded video and one page report due		Jupyter Notbook (Kmeans and DBSCAN);
June 16, 2021	*Evaluation of Clustering Algorithms; *Project guideline; *Class Notes; Refer to Lecture9 video on Canvas
June 18, 2021				Q8(GMM) Q9(DBSCAN)
June 21, 2021	*Density Estimation; *Class Notes; *QA Notes [No Notes]; Refer to Lecture10 video on Canvas Refer to Lecture10-QA video on Canvas [No video]				KDE interactive visualization ; KDE sampling ; KDE SKLearn and sampling ; Jupyter Notebook Kernel Density Example;
June 23, 2021	*Dimension Reduction; *Class Notes; Refer to Lecture11 video on Canvas				Image reconstruction using PCA ; Feature extraction using PCA ; PCA for images ; PCA as linear combination of features ; PCA and Linear Discriminant Analysis ;
June 25, 2021				Q10(Density Estimation) Q11(PCA)
June 28, 2021	*Linear Regression; *Class Notes; *QA Notes [Watch Video]; Refer to Lecture12 video on Canvas Refer to Lecture12-QA video on Canvas	A3 Out	A2 Due		Simple Linear Regression in Matrix Format; Adding Noise to Regression Predictors;
June 30, 2021	*Regularization and Linear Regression; *Class Notes; Refer to Lecture13 video on Canvas Class Video Lecture;
Jul 2, 2021				Q12(Linear reg) Q13(Regularized reg)
Jul 5, 2021	School Break - No class
Jul 7, 2021	*Naïve Bayes and Logistic Regression; *Class Notes; Refer to Lecture14 video on Canvas		Project midpoint report
July 12, 2021	*Neural Networks(Forward pass and Back propagation); *Class Notes; QA Notes [No Notes]; Refer to Lecture15 video on Canvas Refer to Lecture15-QA video on Canvas [No Video]				NN Playground ; Interactive NN initialization ; The role of a hidden layer; Back propagation numerical example; More detailed introduction;
July 14, 2021	*Neural Networks and Deep learning (CNN); *Class Notes; Refer to Lecture16 video on Canvas	A4 Out	A3 Due		CNN Live Demo; A guide to an efficient way to build CNN and optimize its hyper-parameters; Back Propagation in CNN; Transfer learning in CNN; Project Scoring Guidance;
Jul 16, 2021				Q14(Logistic) Q15(NN and CNN)
July 19, 2021	*Decision Tree; *Decision Tree Class Notes; *Ensemble Learning and Random Forest; *Ensemble Learning Class Notes; QA Notes [No notes]; Refer to Lecture17 video on Canvas Refer to Lecture17-QA video on Canvas
July 21, 2021	*Support Vector Machine; *Support Vector Machine Class Notes [Refer to lecture video]; Refer to Lecture18 video on Canvas				KKT and SVM;
Jul 23, 2021				Q16(DT - RF) Q17(SVM)
July 26, 2020	*Kernel Method \ SVM; *Kernel Method \ SVM Class Notes; QA Notes [No Note]; Refer to Lecture19 video on Canvas Refer to Lecture19-QA video on Canvas			A4 Due (can submit till July 28th without any penalty); Final project Due ( can submit till August 2th without any penalty)

Course policies

• Attendance: Our class will be offered in a Synch mode. We will not meet in person and every lecture will be live-streamed and recorded. The recordings will be made available to all students after class time. My on-campus class was mandatory and it helped me and my students A LOT to work with each other for a better environment to facilitate learning. The class is in Synch mode and attendance is required for this class, and it is quite important to join our live-streaming lectures. Trust me it will be fun and you will give me a lot of energy to teach better. The fact that you need to listen to the lectures without fast-forwarding me can help you to learn the materials much better and you will have the chance to ask questions if you are confused anywhere in the lectures. Also, the class attendance will be counted toward your class participation at the end of semester.
• Class deliverables: All class deliverables will be handled via Gradescope except quizzes which will be on Canvas. The time span offered to complete the course objectives is plentiful and deadlines will not be extended under any circumstances. To ensure the class is fair for all students, you will receive zero credit for work submitted after the deadline. Regrade requests should be submitted directly on Gradescope within one week of grade publication. Should you find yourself in an impasse with the TA responsible for your grading, feel free to contact the head TA or course instructor.
• Edstem:
  • Edstem will be the main and only place for the course discussions and announcements. If you have questions, please ask it on Edstem first because 1) other students may have the same question; 2) you will get help much faster.
  • If it’s something you do not like to discuss publicly on Edstem, you can use private messaging on Edstem.
  • Anytime you want to send a private message to just me on Edstem, please make sure to add our HEAD TAs too in case I may miss your message.
  • Edstem GOOD questions
    • I don't understand this part of the lecture, can you explain it to me?
    • This certain part of the hw is not clear to me, would it be possible to explain that more?
    • I have a question about the project ...
    • I found an issue on the website, hw or the lectures, can you clarify ...
    • Any feedback, suggestions, ... would be greatly appreciated.
    • Usually, most of the questions are good.
  • Edstem BAD questions
    • Can you debug my code? [our team will not do that. You need to be specific about your question]
    • Can you find where the problem is in my code?
• Exceptional circumstances: Any request for exceptions to these policies should be made in advance when at all possible. Requests should be due to incapacitating illness, personal emergencies, or similarly serious events. Your request should be accompanied by a supporting letter issued by the Dean of Students.

Diversity and inclusion

Just as machine learning algorithms cannot accomplish complex tasks if trained on datasets of limited variability, our course cannot be successful without appreciating the diversity of our students. In this class we aim to create an environment where all voices are valued, respecting the diversity of gender, sexuality, age, socioeconomic status, ability, ethnicity, race, and culture. We always welcome suggestions that can help us achieve this goal. Additionally, if any of our class scheduled activities conflicts with religious events, please inform the instruction team so that we can make appropriate arrangements for you.

Students with disabilities: your access to this course is extremely important to us. The institute has policies regarding disability accommodation, which are administered through the Office of Disability Services: http://disabilityservices.gatech.edu. Please request your accommodation letter as early in the semester as possible, so that we have adequate time to arrange your approved academic accommodations.

Office hours and questions

Office hours will start on the second week of clases. Please follow the instruction on this Excel Sheet [Office hours] to signup for a 10-minutes slot with one of the TAs. If you require more than ten minutes, please advise the TAs. They’ll return to your BlueJeans meeting once they have completed their appointments with other students. You just need to add your name, question of interest and your BlueJeans meeting link. Please do not change the other part of the Excel Sheet [Office hours]. The TA meetings are designed to be one-on-one. Please do not join another student’s BlueJeans meeting. The sole exception to this policy being discussions about the project, in which your fellow team members can also join. In addition to the one-on-one meetings open office hours with the instructor will be held weekly where you can ask general questions about the topics covered in class. In-person office hours are only available by appointment and will likely be held outdoors, in line with the aforementioned Georgia Tech's and CDC guidelines with respect to preventing the spread of the coronavirus.

Grading

• Assignments (50%)

• There will be four assignments. Each one is designed to improve and test your understanding of the materials. Assignments will have both programming and written analysis components.
• You will need to submit all your assignments using Gradescope. Instructions on how to submit your code and written portions will follow with every assignment. Handwritten solutions WILL NOT BE ACCEPTED and you will not receive credit for a handwritten submission.
• You are required to use Markdown, Latex (watch the tutorial created by our own team [Latex Tutorial] and OverLeaf Latex Example in the Video), or a word processing software to generate your solutions to the written questions. Because handwritten solutions WILL NOT BE ACCEPTED.
• All assignments follow the “no-late” policy. Assignments received after the due date and time will receive zero credit.
• All students are expected to follow the Georgia Tech Academic Honor Code.
• You can easily export your Jupyter Notebook to a Python file and import that to your desired python IDE to debug your code for assignments.
• You are NOT allowed to share any assignment codes or answers with other students. Edstem is the best place to have discussion regarding assignments and course topics. Discussions are just for the better understanding of questions and should not directly answer the questions.
• We have 4 big assignments in total. Visit this course's Canvas site for the assignment documents. See the schedule table above for deliverable due dates. (Topics are subject to change)
  • [12.5%] HW1: Linear Algebra, Probability and Statistics, Maximum Likelihood Estimation, Optimization, Information Theory
  • [12.5%] HW2: KMeans, Expectation Maximization, Gaussian Mixture Model, Clustering Evaluation
  • [12.5%] HW3: Singular Value Decomposition, Principal Component Analysis, Linear Regression, Regularization Naive Bayes
  • [12.5%] HW4: Decision Trees, Random Forest, Support Vector Machine, Neural Networks, CNN



• Project (30%)

• Proposal (5%)

• A project proposal should be written on your GitHub page. It is also a good starter to come up with the first draft of your project.
• You need to provide us the link to your GitHub page. Make sure your GitHub repository is private.
• It should be less than 500 words single spaced. References are not the part of the word count.
• A project proposal should include:
  • Introduction/Background
  • Problem definition
  • Methods
  • Potential results and Discussion (The results and methods may change over the course of this class while you are working on the project and it is fine; that's why it is called research)
  • At least three references (preferably peer reviewed). You need to properly cite the references on your proposal.
• A checkpoint to make sure you are working on a proper machine learning related project.
• Your group needs to submit a presentation of your proposal. Please provide us a public link which includes a 3 minutes recorded video. I found that OBS Studio and GT subscribed Kaltura are good tools to record your screen. Please make your visuals are clearly visible in your video presentation.

• Midterm report (10%)

• A checkpoint to make sure that you have had major progress in your project. You will add information to your project Proposal and turn it into your midterm report.
• You need to provide us the link to your GitHub page. Make sure your GitHub repository is private.
• The midterm report does not have a word count limitation.
• A project midterm report is quite similar to your proposal with the exception of having actual results instead of potential ones:
  • Introduction/Background
  • Problem definition
  • Data Collection
  • Methods
  • Results and Discussion
    • All groups should have their data ready and cleaned at this point
    • We expect to see data pre-processing in your project such as feature selection (Forward or backward feature selection, dimensionality reduction methods such as PCA, Lasso, LDA, .. ), taking care of missing features in your dataset, ...
    • We expect to see at least one supervised or unsupervised method implemented and the results need to be studied in details. For example evaluating your predictive model performance using different metrics (take a look at ML Metrics)
• You do not submit any video for the midterm report.

• Final report (15%)

• You need to provide us the link to your GitHub page. Make sure your GitHub repository is private.
• A final report should include:
  • Introduction/Background
  • Problem definition
  • Data Collection
  • Methods
  • Results and Discussion (We expect to see multiple predictive models and your team need to compare them together and evaluate the results. If your team is working on a Deep learning project, you could finely tune hyperparameters and explain how it could improve the results or you could employ different architectures or methods)
  • Conclusions
• Your group needs to submit a presentation of your final report. Please provide us a public link which includes a 6 to 9 minutes recorded video. I found that OBS Studio and GT subscribed Kaltura are good tools to record your screen. Please make your visuals are clearly visible in your video presentation.



• General project guidance

• Your project will be graded based on the following criteria:
Was the motivation clear?
• What is the problem?
• Why is it important and why we should care?
Were the dataset and approach used effectively?
• How did you get your dataset?
• What are its characteristics (e.g. number of features, # of records, temporal or not, etc.)
• Why do you think your approach can effectively solve your problem?
• What is new in your approach?
Were the experiments, results, and conclusion satisfactory?
• How did you evaluate your approach?
• What are the results?
• How do you compare your method to other methods?
How was the presentation in general?
• Finished on time?
• Effective visualizations? (Are they relevant? Do they help you better understand the project's approaches and ideas?)
• Use of text (Succinct or verbose?)
• Undergrad students can ONLY team up with Undergrad Students, and Grad students can ONLY team up with Grad students. If you are in a Grad students team, you are required to have both unsupervised and supervised learning in your project. I highly recommend Undergrad students to use both unsupervised and supervised learning in your project. However, if you were to pick one, please go with supervised learning.
• In order for you to obtain hands-on experience applying the topics covered in this course, you are expected to complete a term project utilizing real-world data. The project will encompass both unsupervised and supervised learning.
• Each project needs to be completed in a team of five people (you will be forming your team on your own. In case you cannot find a team, we will randomly assign you a team). Team members need to clearly claim their contributions in the project report. Once your teams have been formed and you have selected a topic, you will be assigned a mentor, who will provide you with general guidance on your project. It is important to note that your team will lead the project effort: obtaining the data, researching data-driven approaches to accomplish your project goal and coordinate your own activities. The role of the mentor is solely to advise you, should you find yourself stuck and unable to make progress. We also accept a team of four, if you really cannot find the fifth team member.
• You will create a GitHub page page for your project, which you will use to publish your main deliverables. There will be three deliverables published to your GitHub: a proposal, a midterm checkpoint, and a final report.
• Seminars: To help you conduct your project successfully, We have project seminars where one or two TAs will present their research or industrial projects. Doing so, you will gain a good sense of what it is being done in both Academia and Industry. Besides that, students can ask general questions about their class project and how to improve that in each seminar. Seminars will be streamed online and recorded and they will be published on the course website. Similar to the class lectures, Please ensure that you join to these seminars and get yourself familiar with the practical and real-world application of ML. They will be held on Wednesdays from 11:00 am to 12:00 pm. Timing may change for some of the seminars. We will have Edstem post for each seminar, and its exact time. All the information will be updated for each seminar on the course webiste with a BJ Event link.
• Google colaboratory allows free access to run your Jupyter Notebook. I strongly suggest you use it for your project, especially for teams that are going to employ Deep Learning. Don't forget to take advantage of Google Cloud Platform and AWS Educate as well.



• Quizzes (15%)

• There will be 17 quizzes throughout the semester.
• We will consider your top 10 quizzes' scores. Each quiz will have 1.5% of your final score.
• All quizzes are mandatory to be taken even if they do not count toward your final grade.
• The topic of each quiz will coincide roughly with the content covered in class on that week.
• Quizzes will have a duration of seven-minutes for Undergrad students and six-minutes for Grad students. Each quiz will have five multiple choice questions. They will be available from 12:00 am AOE Friday until 11:59 pm AOE Friday.
• Quizzes measure your understanding of the topics and they will be more conceptual questions.
• Quizzes' answers will be released on Saturdays. Please do not ask any questions about a quiz that you just take on Edstem before we release the answers.
• Quizzes questions are selected randomly from our question bank, which means that students will not receive the same questions for their quiz.



• Class participation (5%)

• Edstem has statistics which give us many measurements regarding how much a student has been involved on Edstem's activities such as viewing posts, answering questions, asking questions and so on. We use this to account for your Class Participation score. At the end of the semester, we will define a minimum and maximum number of involvement considering all the students and your grade will be defined based on that. For Synch classes, lecture participation is considered as a part of the class participation.



• Bonus points (up to 6%)

• About bonus points: Bonus points will be counted to always be beneficial for your final grade. More information on bonus points for assignments will be provided as the semester progresses. If it becomes necessary to curve grades, bonus points will be applied after curving, not before.
• Undergrad and grad: You can obtain up to 3% bonus points would be answering the challenging questions we may have in some of the hws.
• Undergrad: You will notice that we have bonus points for all the hws, where grad students are required to answer those questions, but it will be optional for undergrad students. You will receive up to 3%, if you answer those questions. Note that these are different than the challenging questions. Challenging questions are bonus for both grad and undergrad.
• How does it work? For example, hw 1 may have 30 bonus points, hw 2 may have 20 bonus points and so on. If you receive all the bonus points for all your hws, we will add 3% to your final grade. If you are an undergrad and you answer all the challenging and Grad students questions, you will receive 6%.

COVID-19 Policy