Cpe 695 Syllabus S19

Applied Machine Learning – CPE 695WS Schaefer School of Engineering & Science Spring 2019 Meeting Times: Weekly course materials available on Canvas by each Tuesday Classroom Location: Online Instructor: Shucheng Yu Contact Info: Burchard 412, [email protected], 201-216-8057 Office Hours: Wednesday 3:00pm – 5:00pm Course Web Address: https://sit.instructure.com/courses/32374 Prerequisite(s): College math; familiarity with Python language is a plus. Corequisite(s): None Cross-listed with: N/A COURSE DESCRIPTION Machine learning concerns about improving the performance of computer programs (hence the behavior of computers or systems) via learning from examples or experience. It is a crucial enabler for emerging artificial intelligence and big data applications. This course aims to introduce basic machine learning algorithms and their applications in real problems in addition to fundamental theories underneath these algorithms. Students will have extensive hands-on experiences which prepare them with problem solving capabilities in real applications. LEARNING OBJECTIVES After successful completion of this course, students will be able to… • Understand the basic principles and algorithms of representative machine learning systems including supervised learning, unsupervised learning, batch learning, online learning, model-based learning, and instance-based learning; • Select appropriate machine learning algorithms for real-world tasks; • Implement learning systems and train models with programming languages such as Python; • Choose appropriate performance measurement metrics, tune and evaluate the trained model against the metrics; • Apply related data analytic techniques for data acquisition, cleaning and visualization. FORMAT AND STRUCTURE • This course is comprised of weekly lectures, final group learning projects and presentations, all conducted online. COURSE MATERIALS Textbook(s):

No required textbooks.

Other Readings: 1

Machine Learning, by Tom M. Mitchell, McGraw-Hill Hands-On Machine Learning with Scikit-Learn & TensorFlow, by Aurelien Geron, O’Reilly Publication. Deep Learning, by Ian Goodfellow, Yoshua Bengio, and Aaron Courville, The MIT Press. Materials:

All other materials and slides will be uploaded to course website (Canvas).

COURSE REQUIREMENTS Homework There will be eight (8) homework assignments throughout this course. Each assignment counts for 40 points on average. Each assignment usually dues in TWO (2) weeks after when it is posted on Canvas if programming question(s) are included, or ONE (1) week if no programming question is included. 5 points will be deduced each day after the due date. 320 points possible. Final Project There will be a final team project of the course. Each team comprises three (3) members who can collaborate in person or online. The task of each team member shall be clearly defined. There will be three milestones for the final project: proposal, mid-stage report, and final report & presentation. The proposal will be up to one-page description of the problem of the project and tentative plan. The mid-stage report will be at least three pages. At the end of the semester each team will 1) make a project presentation video and 2) submit a final package (one per team) which includes two files: i) the final report (in one PDF file, 6 pages minimum) and ii) the source code (submitted to Kaggle.com). The project proposal will be 20 points, the mid-stage report 40 points; the final presentation video counts for 60 points. The writing of the final report will be 60 points, and the overall methodology and quality of the project counts for 180 points. 360 points possible. Exams There will be a mid-term exam and a final exam for this course, both online. The mid-term exam counts for 200 points, and the final exam will be 120 points. The final exam is NOT cumulative. There will be an online review lecture one week before each exam. There is NO makeup exams. Excused absence from any exam shall seek consent from the instructor prior to the exam day. 320 points possible. GRADING PROCEDURES Grades will be based on: Homework (32 %) Final Project (36 %) Mid-term Exam (20 %) Final Exam (12 %) Total 100%

320 points 360 points 200 points 120 points 1000 points

ACADEMIC INTEGRITY Graduate Student Code of Academic Integrity All Stevens graduate students promise to be fully truthful and avoid dishonesty, fraud, misrepresentation, and deceit of any type in relation to their academic work. A student’s submission of work for academic credit indicates that the work is the student's own. All outside assistance must be acknowledged. Any student who violates this code or who knowingly assists another student in violating this code shall be subject to discipline. All graduate students are bound to the Graduate Student Code of Academic Integrity by enrollment in graduate coursework at Stevens. It is the responsibility of each graduate student to understand and adhere to the Graduate Student Code of Academic Integrity. More information including types of violations, the 2

process for handling perceived violations, www.stevens.edu/provost/graduate-academics.

and

types

of

sanctions

can

be

found

at

EXAM ROOM CONDITIONS The following procedures apply to quizzes and exams for this course. As the instructor, I reserve the right to modify any conditions set forth below by printing revised Exam Room Conditions on the quiz or exam. 1. Students may use the following devices during quizzes and exams. Any electronic devices that are not mentioned in the list below are not permitted. Device Laptops Cell Phones Tablets Smart Watches Google Glass Other (non-programmable calculator)

Permitted? Yes No X X X X X X

2. Students may use the following materials during quizzes and exams. Any materials that are not mentioned in the list below are not permitted.

Material Handwritten Notes Typed Notes Conditions: one A4 sheet (front and back) Textbooks

Permitted ? Yes No X X X

Readings

X

3. Students are not allowed to work with or talk to other students during quizzes and/or exams. LEARNING ACCOMODATIONS Stevens Institute of Technology is dedicated to providing appropriate accommodations to students with documented disabilities. The Office of Disability Services (ODS) works with undergraduate and graduate students with learning disabilities, attention deficit-hyperactivity disorders, physical disabilities, sensory impairments, psychiatric disorders, and other such disabilities in order to help students achieve their academic and personal potential. They facilitate equal access to the educational programs and opportunities offered at Stevens and coordinate reasonable accommodations for eligible students. These services are designed to encourage independence and self-advocacy with support from the ODS staff. The ODS staff will facilitate the provision of accommodations on a case-by-case basis. Disability Services Confidentiality Policy Student Disability Files are kept separate from academic files and are stored in a secure location within the Office of Disability Services. The Family Educational Rights Privacy Act (FERPA, 20 U.S.C. 1232g; 34CFR, Part 99) regulates disclosure of disability documentation and records maintained by Stevens Disability Services. According to this act, prior written consent by the student is required before our 3

Disability Services office may release disability documentation or records to anyone. An exception is made in unusual circumstances, such as the case of health and safety emergencies. For more information about Disability Services and the process to receive accommodations, visit https://www.stevens.edu/office-disability-services. If you have any questions please contact: Phillip Gehman, the Director of Disability Services Coordinator at Stevens Institute of Technology at [email protected] or by phone (201) 216-3748. INCLUSIVITY Name and Pronoun Usage As this course includes group work and in-class discussion, it is vitally important for us to create an educational environment of inclusion and mutual respect. This includes the ability for all students to have their chosen gender pronoun(s) and chosen name affirmed. If the class roster does not align with your name and/or pronouns, please inform the instructor of the necessary changes. Inclusion Statement Stevens Institute of Technology believes that diversity and inclusiveness are essential to excellence in academic discourse and innovation. In this class, the perspective of people of all races, ethnicities, gender expressions and gender identities, religions, sexual orientations, disabilities, socioeconomic backgrounds, and nationalities will be respected and viewed as a resource and benefit throughout the semester. Suggestions to further diversify class materials and assignments are encouraged. If any course meetings conflict with your religious events, please do not hesitate to reach out to your instructor to make alternative arrangements. You are expected to treat your instructor and all other participants in the course with courtesy and respect. Disrespectful conduct and harassing statements will not be tolerated and may result in disciplinary actions. TENTATIVE COURSE SCHEDULE The following is a tentative course schedule. Any changes to this schedule will be communicated to you 1) via class lecture and/or 2) via email. The Canvas shell for this course will always be kept up-to-date so you can always reference the “Assignments” tab for accurate due dates. Weeks #1: Jan. 22 - 25 #2: Jan. 28 – Feb.1 #3: Feb. 4 - 8 #4: Feb. 11 - 15 #5: Feb. 18 - 22 #6: Feb. 25 – Mar.1

Topic(s) Introduction & Tools

Linear Regression Logistic Regression; Concept Learning Decision Tree (1) Decision Tree (2); Ensemble Learning; Random Forests Support Vector Machine

Readings Required: None Optional: Ch.1 of Mitchell Book Ch.1 & 2 of Geron Book Required: None Optional: Ch.4 of Geron Book Ch.3 of Bishop Book Required: Ch.2 of Mitchell Book Optional: Ch.4 of Geron Book Ch.4 of Bishop Book Required: Ch.3 of Mitchell Book Optional: Ch.6 of Geron Book

Assignment

Homework #1 due Homework #2 due

Required: Ch.3 of Mitchell Book Optional: Ch.7 of Geron Book Optional: Ch.5 of Geron Book

4

Homework #3 due

#7: Mar. 4 - 8

Bayesian Learning; mid-term review

#8: Mar. 11

Mid-term exam

#9: Mar. 18 - 22

Spring Break

#10: Mar. 25 - 29

Artificial Neural Networks (1)

#11: Apr. 1 - 5

Artificial Neural Networks (2);

#12: Apr. 8 - 12

Evaluating Hypothesis; Dimensionality Reduction Instance-Based Learning; Genetic Algorithms Deep Belief Network; Convolutional Neural Networks; Autoencoders Recurrent neural network; LSTM Network; Final review

#13: Apr. 15 - 19 #14: Apr. 22 - 26 #15: Apr. 29 – May 3

#16: May 6

Homework #4 due Required: Ch.6 of Mitchell Book

Required: Ch.4 of Mitchell Book Optional: Ch.10 of Geron Book Required: Ch.4 of Mitchell Book Optional: Ch.10 of Geron Book Required: Ch.5 of Mitchell Book Optional: Ch.8 of Geron Book Required: Ch.8 & 9 of Mitchell Book

Homework #5 due Final project proposal due

Homework #6 due Homework #7 due Mid-stage project report due

Optional: Ch.13 of Geron Book Ch.15 of Geron Book Optional: Ch.14 of Geron Book

Homework #8 due

Final Exam

May 12 Final project report due

5