Spring (24)
This is a problem-driven course: we start by exploring how to use Large Language Models to solve concrete problems in the world and gradually progress toward a more technical understanding as the term advances.
Office Hours
| Day | Time | Location |
|---|---|---|
| Tuesday | 3:30 – 4:30 PM | Lewis and Gaines Center for Inclusion and Equity |
| Thursday | 3:30 – 4:30 PM | DMF 461 |
Materials
- Prerequisite: Linear Algebra and Multivariable Calculus
- Some notes
- ChatGPT Cheat Sheet (thanks to Uma Shama for sharing this)
- Supplementary AI materials
| Free textbook | Introduction to Machine Learning by Etienne Bernard |
| Mathematica (Wolfram Cloud) | Go to https://www.wolframcloud.com/ Use your BSU credentials to sign up You will get to your notebook on Cloud. |
| ChatGPT +Wolfram | ChatGPT with Wolfram Plugin (need an API) |
| Machine learning basics | |
| What is ChatGPT doing? |
The structure of this course is two-fold. For the first part of the course, each student will identify personal directions which they aim to develop and how Large Language Models can assist in this personal development task. The second part of the course will be more technical: we will learn together how mathematical objects such as neural networks, the technology behind ChatGPT are trained and as a result are able to learn.
About Grades
Final Grade = (Homework Score x 20%) + (Attendance Score x 10%) + (Midterm Score x 30%) + (Final Exam Score x 40%)| Feature | Description |
|---|---|
| Compute your grades in your browser | A tool to calculate grades directly in the web browser |
| Compute your grades with Mathematica | A method to calculate grades using Mathematica software |
About Homework
| Item | Description |
|---|---|
| Homework Platform | Blackboard |
| Access Instructions | Use BSU credentials |
Chair Scheduler (trained by Vignon Oussa with the help of Judi Morin)
Chair Scheduler is a specialized version of ChatGPT designed to assist with scheduling tasks, particularly in academic settings or similar environments. Its capabilities are centered around interpreting, organizing, and suggesting schedules based on various inputs such as faculty availability, course assignments, and specific time blocks. It may assist with:
- Schedule Drafting
- Conflict Detection
- Data Representation
- Document Interpretation
- File Creation
- Interactive Assistant
Sketching the Semester
| Month | Topics | Details |
|---|---|---|
| January | Learning how to use Large Language Models (ChatGPT) | Project 1 |
| February | Large Language Models (ChatGPT) | Project 2 |
| March | Machine Learning | Machine Learning |
| April | Neural Networks |
(1) Neural Network (a) Notes on shallow and deep neural networks (2) Solutions of Midterm (3) Project 4: Exploring the world of neural networks
|
| May | Final Exam | Wednesday, May 1: 11 a.m. – 1 p.m. |