EE269 - Signal Processing and Quantization for Machine LearningAnnouncements
Course descriptionThis course introduces key signal processing and quantization concepts for modern machine learning and AI. Students learn techniques for capturing, processing, and classifying signals, tracing the roots of quantization in signal processing and its role in generative AI. Topics include signal models, vector spaces, Fourier and time-frequency analysis, Z-transforms, filters, wavelets, autoregression, image and video processing, matrix decompositions, compressed sensing, deep learning, and mixed-precision quantization, with applications ranging from adaptive filters to large language models and other generative AI systems. This class will culminate in a final project. Prerequisites:Exposure to signals and systems (EE 102A and EE 102B or equivalent), basic probability (EE 178 or equivalent), basic programming skills (Matlab), familiarity with linear algebra (EE 103 and EE 178 are recommended). Instructor: Mert Pilanci, pilanci@stanford.edu |