Tentative Lecture Schedule

This schedule is only tentaive and may change drastically during the quarter. The exam date, however, will not change.
Slides will usually be posted on the course home page before each lecture. Since lectures may differ substantially from the textbook, attending live or watching recordings promptly is the best way to stay on track.
Guest lectures are not part of the exam syllabus, but they will be central to the extra credit activities.

Weekly Topics

Week	Focus
1-2	Foundations: deep learning, computational graphs, autodiff, and ML frameworks
3	GPUs, CUDA, and collective communication
4	Graph and memory optimizations
4	Guest lecture (TBD): GPU and hardware trends
5	Parallelism in ML systems
6	Transformers, LLMs, and diffusion models
6	Guest lecture (TBD): LLM systems
7	LLM training: FlashAttention and related efficiency techniques
8	LLM inference and serving: paged attention, continuous batching, and speculative decoding
9	Guest lecture (TBD): agents
9	Scaling laws, test-time compute, and reasoning
10	LLM + X: RAG, search, multimodality, and related directions
10	Guest lecture (TBD): TBD
10	Final exam review
11	Final exam