Hardware Accelerators for Machine Learning (Spring 2023)

ECE 618

Coure Inforamtion

Instructor Dr. Weiwen Jiang
E-Mail wjiang8@gmu.edu
Phone (703)993-5083
Lecture Time Monday 19:20 - 22:00
Location Room 131, Innovation Hall
Office Hour Monday 16:00 - 17:00
Office Room 3247, Nguyen Engineering Building
Zoom https://go.gmu.edu/HA4ML23

Course Materials

Course materials will be posted before or after the class. No formal textbook is required.

(ECE 511 or CS 465) and (ECE 527 or ECE 554 or CS 580 or CS 688) or permission of instructor.

Course Description

This course covers the hardware design principles to deploy different machine learning algorithms. The emphasis is on understanding the fundamentals of machine learning and hardware architectures and determine plausible methods to bridge them. Topics include precision scaling, in-memory computing, hyperdimensional computing, architectural modifications, GPUs and vector architectures, quantum computing as well as recent hardware programming tools such as Xilinx AI Vitis, Xilinx HLS, and IBM Qiskit.

Tools for Lab

Schedule and Documents

[Syllabi]

W Date Topic Documents Note
Session I: Classical Computing Accelerators for Machine Learning
1 Jan. 23 Course Information & Machine Learning and FPGA Accelerator Recap
2 Jan. 30 Domain Specific Computing
3 Feb. 6 Vector Architectures, GPU Architectures, and Benchmarking Metrics
Session II: Novel Post-Moore Computing Accelerators for ML
4 Feb. 13 FPGA Accelerator Designs
5 Feb. 20 In-Memory Computing Accelerator Design
6 Feb. 27 Hyperdimensional Computing Accelerators (1)
7 Mar. 06 Hyperdimensional Computing Accelerators (2) and Midterm Review
8 Mar. 20 Midterm Exam
Session III: ML Accelerators in Quantum Computing
9 Mar. 27 Project Proposal & Single Qubit System
10 Apr. 03 Quantum Neural Network Accelerators
11 Apr. 10 Hands-on Accelerator Design (1)
12 Apr. 17 Project Overview
13 Apr. 24 Hands-on Accelerator Design (2)
14 May 01 Project Presentations

Readings and Tutorial

W Date Reading (R) & Paper (P) & Tutorial (TT)
1 Jan. 23 [HLS Doc] [HLS Youtube]