Foundation Models: Principles and Practice

COM SCI C160F - Foundation Models: Principles and Practice

<p>Lecture, four hours; discussion, two hours; outside study, six hours. Requisite: course M146. Discussion of advanced topics and state-of-the-art research in training foundation models, including large language models (LLMs) and large vision language models (LVLMs). Topics include pretraining and post-training foundation models, safety alignment, training reasoning LLMs and VLMs, and synthetic data generation for foundation models. Concurrently scheduled with course C160F. Letter grading.</p>

Computer Science

Henry Samueli School of Engineering and Applied Science

Container 1

COM SCI C260F -  Foundation Models: Principles and Practice

 Foundation Models: Principles and Practice

Position students to contribute to topics including large-scale pretraining and adaptation techniques (e.g., fine-tuning, prompting, and alignment), reasoning, and synthetic data generation.

 CS students to design, analyze, and responsibly deploy these systems, while also positioning them to contribute to the next generation of AI methods rather than treating these models as opaque black boxes. 

General Catalog

Foundation Models: Principles and Practice

COM SCI C260F

4 units