Class Attendance and Engagement
In-person attendance at class is generally expected. Class time has been reserved for our coming together and provides an opportunity to focus, put away distractions, get to know each other, and ask questions. More than attend, you are expected to be engaged during class. You are welcome to use electronic devices to take notes, consider code, or otherwise engage in materials related to the course. You should not use electronic devices for other purposes during class that are distracting to yourself and your peers, including but not limited to social media, instant messaging, or video streaming.
You may sometimes need to miss class due to illness, emergencies, or other obligations. Absence will not be punished / your grade will not be penalized. If you must miss class, you are responsible for reviewing the class materials.
References
Lectures will generally introduce new concepts, and you are not expected to read prior to lectures. However, you should expect to read to deepen your understanding of a topic after lecture, and should bring any outstanding questions with you to class, ask online, or in office hours. You will get the most out of the course depending on what you put in, and the course references are rich sources of information beyond the minimum that we can introduce during lecture. Specific suggestions for readings are marked on the schedule under references.
Graded Items and Course Grades
The course is organized into four units, each of which weights equally, 25% of the course grade. Each unit has two graded items: A homework assignment completed outside of class and an in-class quiz. In homework assignments you will work with concepts from class in code, described in more detail on the assignments page. The quizzes will be 35-40 minutes long written, in-class, closed book and closed notes.
Each quiz also has a corresponding redux opportunity. The redux is an optional quiz that will be comparable in content and structure. Your quiz grade will be the better of the original and the redux. The redux thus serves as the makeup if you are unable to take the original quiz, as well as an opportunity to improve your score if you are unhappy with your original quiz score.
There is no midterm or final exam scheduled for the Spring 2025 semester.
Numerical course grades will be computed based on the following weighted average of categories.
| Unit | Total Weight | Graded Items | Item Weight | Notes |
|---|---|---|---|---|
| 1: Introduction to Machine Learning | 25% | Assignment 1 | 13% | |
| Quiz 1 | 12% | Better of Q1 and Q1R | ||
| 2: Artificial Neural Networks and Image Recognition | 25% | Assignment 2 | 13% | |
| Quiz 2 | 12% | Better of Q2 and Q2R | ||
| 3: Transformers and Language Models | 25% | Assignment 3 | 13% | |
| Quiz 3 | 12% | Better of Q3 and Q3R | ||
| 4: Reinforcement Learning and Behavior | 25% | Assignment 4 | 13% | |
| Quiz 4 | 12% | Better of Q4 and Q4R |
Course letter grades will be assigned based on numerical course grades according to the following threshold function. There is no statistical curve, meaning your grade does not depend on the grades of your peers. There is no rounding rounding up; a numerical course grade must be strictly greater than or equal to the given threshold to earn the given letter.
| Letter | Needs at Least | Letter | Needs at Least | Letter | Needs at Least |
|---|---|---|---|---|---|
| A+ | 97 | B | 83 | C- | 70 |
| A | 93 | B- | 80 | D+ | 67 |
| A- | 90 | C+ | 77 | D | 63 |
| B+ | 87 | C | 73 | D- | 60 |
Collaboration and Authenticity
Collaboration is an encouraged part of the course, but only within boundaries that will ensure your learning and maintain academic integrity. The homework assignments can be completed in groups of size 2 (that is, you can work with a partner). This is optional: you can by yourself if you like, and we do not assign mandatory partners. You can switch partners between assignments, but once you begin working with a partner for a particular assignment you may not change for that same assignment. Beyond your partner, you may discuss ideas and study with other students in small groups of up to 6. You may not, however, share code outside of your partner.
It is expected that you will use resources on the web to help you study and complete your labs, including official documentation, examples, and discussions. However, the work you submit should be authentically your own. To qualify as an author of your own work, you should be the primary drafter, meaning you should never copy text/code without direct citation and attribution of sources, even if you paraphrase it in your own words.
AI – Large Language Models (LLMs)
You may choose to use LLMs such as ChatGPT in this course but should do so according to the following guidelines of appropriate use. These guidelines are inspired by those of the NeurIPS author instructions, a leading venue for the publication of machine learning research, adapted to an educational context.
- You are ultimately responsible for the entire content of any submitted work, meaning it is your responsibility to ensure the integrity, correctness, and originality of your work.
- You should never characterize work from an LLM as your own without attribution or discussion of the method of your usage. If you generated some text/code with an LLM then reviewed and edited it, you should note as much.
- In an educational context where the goal is to learn, you should endeavor to use LLMs to supplement and enhance your learning rather than to substitute for your learning.
- Do: get help but build independence, ask clarifying questions, ask to explain things in a different way, ask for help with interpreting parameters or error messages, ask for help understanding documentation or API usage, examine sources and try suggestions for yourself, rely on your own judgement, ensure that you understand your own work, be skeptical of the answers you receive.
- Do not: blindly trust LLM outputs, use LLM output to avoid trying something you want to learn yourself, misrepresent LLM work as your own without attribution or discussion, become totally reliant on LLMs.
Disability Accommodation
Duke University is committed to providing equal access to students with documented disabilities. Students with disabilities may contact the Student Disability Access Office (SDAO) to ensure your access to this course and to the program. There you can engage in a confidential conversation about the process for requesting reasonable accommodations both in the classroom and in clinical settings. Students are encouraged to register with the SDAO as soon as they begin the program. Please note that accommodations are not provided retroactively. More information can be found online at access.duke.edu or by contacting SDAO at 919-668-1267, SDAO@duke.edu.
Academic Integrity
The Duke Community Standard applies at all time:.
Duke University is a community dedicated to scholarship, leadership, and service and to the principles of honesty, fairness, respect, and accountability. Citizens of this community commit to reflect upon and uphold these principles in all academic and non-academic endeavors, and to protect and promote a culture of integrity. To uphold the Duke Community Standard:
- I will not lie, cheat, or steal in my academic endeavors
- I will conduct myself honorably in all my endeavors
- I will act if the Standard is compromised
