11-767 On-Device Machine Learning

12 units, Carnegie Mellon University

OnDevice Machine Learning is a projectbased course covering how to build, train, and deploy models that can run on lowpower devices e.g. smartphones, refrigerators, and mobile robots. The course will cover advances in distillation, quantization, weight imprinting, power calculation and more. Every week we will discuss a new research paper and area in this space one day, and have a lab workinggroup the second. Specifically, students will be provided with lowpower compute hardware e.g. SBCs and inference accelerators in addition to sensors e.g. microphones, cameras, and robotics for their course project. The project will involve three components for building lowpower multimodal models 1 inference 2 performing trainingupdates for interactive ML, and 3 maximizing power. The more that can be performed on a device, the more privacy preserving and mobile the solution is. For each stage of the course project, the final model produced will have an mAh budget equivalent to one full charge of a smartphone battery

Prerequisites: none listed

Courseweave needs JavaScript to show sections, workloads and planning tools.