You can find useful resources for this course below
TA :
Matthew Ho mah048@ucsd.edu
Frederick Pi, xpi@ucsd.edu
Alex Rojas a5rojas@ucsd.edu
Office Hour:
Name |Time | Date | Location
Frederick | 3:00 - 4:00 pm | Thursday | https://ucsd.zoom.us/j/5626625274
Alex | 10:00 am - 11:00 am | Wednesday | CSE 4217
Matthew | 4:00 pm - 5:00 pm | Friday | CSE 4142
In this course, you will learn the fundamentals of deep learning. Part of the learning will be through in-class lectures and take-home assignments, but you will really gain hands-on experience by participating in Deep Learning Competitions.
You will participate in a deep learning competition via Kaggle in small groups, normally 1-4 people. The project contributes to 45% of the course credits:
10 % milestone report
15 % final report
10 % final presentation
10 % competition ranking
In the end, you will write a 4 page report about your project.
This competition is a scaled down version of nanogpt speedruns (Keller repo, Prime Intellect Leaderboard). Participants will compete to train a tiny language model to achieve the lowest perplexity on a hidden test set in a scale-constrained environment.
Compute: UC San Diego provides Data Science/Machine Learning platform for this course. You can log in with your Active Directory ID.
In addition, you can apply for student cloud computing credits at Google Cloud and Amazon AWS.