DL
Deep Learning (CMPSCI 4390/5390)
Spring 2025 - Hybrid (online and in-person)
Instructor & Contact Information
- Instructor (Facilitator): Badri Adhikari
- Email:
adhikarib@umsl.edu - Class meets: Wed 6:55 - 8:10 PM at ESH 106
- Office hours: Wed 5:30 to 6:45 PM (in person). Please email me in advance to avoid any potential waiting.
Welcome to the Deep Learning course! I’m thrilled to have you here as we embark on this exciting journey together. Each spring, this course gives me the opportunity to combine my passion for Deep Learning with the joy of guiding others to explore the topic. As Deep Learning is also a core focus of my research, I look forward to learning and growing alongside you this semester!
Your success in this class is very important to me. If you require official accommodations, please know that you have the right to have these needs met. If any aspect of this course creates challenges or barriers to your learning, I encourage you to let me know as soon as possible. Together, we’ll work to develop strategies that address your needs while ensuring the course requirements are met.
“Right now, it may seem hard to believe that AI could have a large impact on our world because it isn’t yet widely deployed—much as, back in 1995, it would have been difficult to believe in the future impact of the internet. Back then, most people didn’t see how the internet was relevant to them and how it was going to change their lives. The same is true for deep learning and AI today. But make no mistake: AI is coming.” - François Chollet (author of our textbook).
“The ultimate aim (of artificial intelligence and deep learning) is to use these general-purpose technologies and apply them to all sorts of important real-world problems.” - Demmis Hasssabis (CEO of Deepmind).
My Teaching Philosophy: In my classes, I see students as “learners” and myself as a “facilitator.” Many learners perceive computer science and technology as a challenging discipline because it is constantly evolving. To help my learners, I recommend a strategy for mastering the fundamentals: an iterative process of reading, analyzing, and coding. Unfortunately, many learners tend to favor either analyzing or coding, but not both. An effective computer science course should therefore strike a balance between (a) theoretical knowledge for understanding computer technology and (b) practical implementation skills to test and execute the theories and algorithms. Hence, I design course content and assignments to help learners improve both their analytical and programming skills. Learners with extensive programming experience may find this balance slightly easier to manage, but they will still have opportunities to expand their theoretical foundation. For those who may not yet consider themselves expert programmers, such a course can help them develop practical implementation skills.
About the Course
Course Name
Introduction to Deep Learning (CMPSCI 4390) / Deep Learning (CMPSCI 5390)
Course Description in UMSL Catalog
This course reviews a typical machine learning recipe, mathematical foundations for deep learning, and provides an introduction to deep learning. Topics include dense neural networks, convolutional neural networks, and recurrent neural networks. The course will cover building, training, and using deep neural networks for solving various machine learning problems like image classification and protein contact prediction. Credit cannot be granted for both CMP SCI 4390 and CMP SCI 5390. [3 credit units].
Course Prerequisites
- For 4390 (for undergraduate students): CMPSCI 3130 (Design and Analysis of Algorithms)
- For 5390 (for graduate students): Graduate Standing in CS
Course Overview
This course covers a typical machine learning workflow, the computational foundations of deep learning, and an introduction to dense neural networks. It focuses on building, training, and evaluating deep convolutional neural networks to solve various machine learning problems, particularly those involving image data. By the end of the course, you will be able to identify the types of problems best suited for deep learning algorithms, as well as those that are not, and develop your own deep learning applications. Additionally, you will gain insight into major technology trends in deep learning and understand what sets it apart from traditional machine learning.
Learning Outcomes
Upon completing the course learners will be able to:
- Use Python, Numpy and Keras to design, train, and evaluate convolutional neural network models
- Write Numpy and Tensorflow programs for building deep learning applications
- Learn computational foundations of machine learning and deep learning
- Learn the major technology trends in deep learning such as convolutional neural networks and residual neural networks
- Understand what makes deep learning different and powerful compared to traditional machine learning practices
- Understand the main parameters and hyper-parameters in a deep neural network
- Design, develop, train, debug, and evaluate deep neural networks
- Learn how to transfer the knowledge in pre-trained deep learning models to build models for new datasets
Required Textbook
Second Edition 🔥 of “Deep Learning with Python” by François Chollet by Manning; associated Notebooks are freely accessible at GitHub.
Course Topics
- The mathematical building blocks of neural networks (Chapter 2)
- Introduction to Keras and TensorFlow (Chapter 3)
- Getting started with neural networks: Classification and regression (Chapter 4)
- Fundamentals of machine learning (Chapter 5)
- The universal workflow of machine learning (Chapter 6)
- Working with Keras: A deep dive (Chapter 7)
- Introduction to deep learning for computer vision (Chapter 8)
- Advanced deep learning for computer vision (Chapter 9)
Academic Honesty
Any deceptive use of AI technologies or academic dishonesty in this class will result in a failing grade for the semester. Additionally, the case will be referred to the Office of Academic Integrity and Student Conduct for further disciplinary action, regardless of the severity of the violation. To avoid academic integrity issues and maximize your learning in this course, here are some proactive steps you can take:
- 🙅♂️ Do your work independently. Do not use another student’s assignment or a solution found on the internet to complete your work. While discussing course material with peers is fine, ensure that the assignments are completed individually—not collaboratively or in a group. If you have any questions about the lessons or assignments, please feel free to reach out to me, and I will guide you in the right direction.
- ✍️ Write in your own words. In assignments where you are expected to explain or summarize your understanding (e.g., writing video summaries), write the answers in your own words.
- 📝 Understand plagiarism. Please read UMSL’s academic integrity policy to familiarize yourself with the rules and avoid plagiarism.
- 🎓 Educate yourself about plagiarism. If this is your first semester at UMSL, or if you’re unsure about what “plagiarism” entails, I recommend watching this lecture on plagiarism available on Coursera.
- 🔒 Respect academic integrity in assessments. Do not print, copy, or share quiz or exam questions outside the class under any circumstances.
- 🔍 Cite your sources and explain your process. All sources must be clearly cited in your submissions and reports. Do not copy and paste content from the internet without proper citation. If you didn’t write it or create it entirely on your own, ensure you cite your sources or clearly explain the process you followed.
Programming Language
Python3 is the programming language for this course; you are expected to use Python3 for all your classroom activities, assignments, and projects. You are welcome to use Google colab or your own hosted Jupyter Notebook to run your programs.
Assignments, Quizzes, and Exams
Assignments
This course will have several types of assignments: chapter readings, regular course assignments, project-related assignments, and drawing concept maps as chapter summaries. All assignments should be submitted via Canvas. I ignore any assignments submitted to me via email. If you need help finding the right place to submit in Canvas, ask me by sending an email instead of emailing me your assignments.
Quizzes
Many classes will begin with a short quiz (5 to 10 minutes). Mostly, these will be multiple-choice questions or numericals. Questions will be mainly from the book chapter readings, classroom lectures, assignments, and classroom activities.
Feedback and Grading Timeline
You will receive quiz grades within a week; others may take longer. Assignments may take longer to grade during certain weeks when I travel to conferences. However, you are welcome to email me and request a quicker grading for a particular submission.
Exams
There will be no traditional comprehensive exams in this course. Instead, the final exam will be an oral presentation (viva) where you will discuss and explain your final project.
Showing Your Process
In this course, some assignments may require you not only to submit a final product but also to demonstrate the process you followed to create it. For example, you might be required to write your project report entirely in Google Docs followed by the use of process-revealing Google Chrome extensions such as Process Feedback. Assignments that have such a requirement will include additional instructions.
Attendance, Due dates, Late policy, and Discussion Board Assignments
- Attendance will be recorded in several classes. If you miss more than 80% of the classes, you will fail the course.
- All assignments are due by midnight on the day of the class. If you require special accommodations (e.g., medical reasons), please email me as soon as possible.
- Assignments and project phases must be submitted by their respective due dates (see Canvas for details).
- You may request a two-day extension on up to two submissions (an assignments, project, or everything that is due on that day). If you email me a few hours before the deadline and don’t receive an immediate response, you can assume the extension is granted, provided you haven’t used your allotted extensions.
- After using your two extensions, late submissions will receive no points.
- For graded discussion boards, you must submit your answer before viewing others’ responses. Submitting a placeholder or dummy answer before posting your actual response is not allowed. Your first submission will be graded, even if it is incomplete or a dummy answer.
Submitting Assignments
Please follow these instructions when submitting assignments via Canvas. Non-compliance may result in a grade deduction of up to 100%.
- Do not submit links alone (e.g., Google Docs or Google Colab). Submit your work in universally accessible formats like PDF, DOCX, or HTML. Links can be included as supporting materials but are not sufficient on their own.
- If you are submitting multiple files in a compressed format (e.g., ZIP), ensure that key documents (e.g., PDFs) are also uploaded separately. This makes it easier for me or the TA to grade your work.
- For submissions with multiple images, combine them into a single document (PDF or Word document).
- You can choose any standard format for citations, such as MLA and APA.
Zoom Meetings
If we ever need to Zoom (for class meetings or office hours), say due to severe weather, you are required to keep your video on.
Grade Composition (tentative)
| Scoring aspect | Total points (%) |
|---|---|
| Weekly quizzes | 50 |
| Assignments | 25 |
| Semester Project | 25 |
Note: You are required to submit the course evaluation survey at the end of the semester to receive your final grade.
Grading Scheme
| Points (%) | Grade | Points (%) | Grade | |
|---|---|---|---|---|
| 94 to 100 | A | 74 to 77 | C | |
| 90 to 94 | A- | 70 to 74 | C- | |
| 87 to 90 | B+ | 67 to 70 | D+ | |
| 84 to 87 | B | 64 to 67 | D | |
| 80 to 84 | B- | 61 to 64 | D- | |
| 77 to 80 | C+ | 0 to 61 | F |
Additional Policies and Resources
- Here are additional course policies and student resources.
- Notebooks associated with the textbook.
- Tutorial: Designing a Concept Map for Maximum Learning
- Tutorial: BA’s Overleaf Template for Writing a Project Report
- Tutorial: How to Prepare for and do an Effective Slide Presentation?
Recommended Study Strategy
This is an intensive course; you should plan to dedicate a minimum of 10 hours per week to it. In addition to attending lectures, you will need to read assigned book chapters, watch videos, and complete assignments and projects. Each week, aim to begin reading the assigned chapters as soon as they are posted. Ideally, complete the reading assignments the day after class.
I encourage you to connect with your classmates and form study groups to discuss assignment questions and solution strategies. Be sure to monitor Canvas for assignment deadlines. To share your findings and engage with others, use the classroom discussion board.
To support you, I will send announcements frequently to remind you of important deadlines and introduce upcoming topics.
Note: Any important changes to this syllabus will be promptly communicated to the class.