As the correlation of data gaining importance in many domains, compared to the traditional studies on attributes, high-dimensional tensors are becoming an ever more important object to represent data and analyze its inherit properties. Its applications cover machine/deep learning, quantum chemistry/physics, quantum circuit simulation, social networks, and healthcare to name a few.
In this course, we will explore the state-of-the-art innovations to accelerate tensor computations, including sparsity, data structure, parallel computing, compression, compilation methods, and architecture design. A mixed pattern of lectures and student presentations will be used. Paper reading, presentation, and an embryonic project are expected.
Instructor: Jiajia Li (http://jiajiali.org). Email: [email protected] Office: EB2 3260
Class Schedule: F 9:35am-12:20pm ET @ 1229 Engineering Building 2
Office hours: Thursday 10-11am ET (other time per request) @ Zoom/Office (https://ncsu.zoom.us/j/99725615611?pwd=UHhxT1h1Ym90ZTlYNFk4TXUwYXJLZz09) (Please drop me an email before coming to the office hour.)
Message Center and Online Discussion: Moodle
Course Videos: https://ncsu.hosted.panopto.com/Panopto/Pages/Sessions/List.aspx?folderID=8ce4fba2-505b-474e-9546-b05500aba61a
Contact: Please do not hesitate to contact me by email.
Grader: Siddarth Menon ([email protected])
CSC 505: Design and Analysis of Algorithms
CSC 548: Parallel Systems
Note: Not much pre-requisites are needed for this class. I will give an introduction to deep neural networks and basic parallel knowledge at the first couple of classes.