Systems for Machine Learning

Description from DukeHub

Principles of designing and implementing systems that enable running machine learning workloads efficiently on modern accelerators (e.g., GPUs). Topics include automated differentiation and backpropagation, machine learning compilers, GPU kernel design and optimization, distributed training, efficient inference techniques, emerging model architectures, parameter-efficient fine-tuning, and reinforcement learning. Prerequisites: Computer Science 310 or 510 or 512; and Computer Science 372 or 527 or 571 or 572 or 574 or 676.

Skills

• programmingdepartment: COMPSCI
• ai and machine learningtitle: "machine learning"
• designdescription: "designing" and 1 more

AI-resilient: 0 of 5 0/5

• No: Physical, lab or field work
• No: Live work with people
• No: Judgment someone is accountable for
• No: A craft you practice with your hands or body
• No: Small, in-person format

The checks lean on Microsoft Research, Working with AI (2025), which found generative AI applies least to physical work and hands-on work with people. A signal, not a promise. How the score works

More in COMPSCI

• Introduction to Computer Science COMPSCI 101L
• Data Structures and Algorithms COMPSCI 201
• Introduction to Computer Systems COMPSCI 210D
• Everything Data COMPSCI 216
• Discrete Math for Computer Science COMPSCI 230
• Computer Architecture COMPSCI 250D

All COMPSCI courses · All of Graduate School