Artificial Intelligence Hardware

David Yeh, Director

Creating new highly-efficient AI platforms to enable neuro-inspired, cognitive and learning abilities which will be required to address the vast range of future data types and workloads as intelligence is enabled from edge devices to the cloud.


This research program highlights the importance of hardware in pushing the frontiers of artificial intelligence across a broad spectrum of applications from the edge to the cloud. The program has roots in two previous efforts: System Level Design (SLD) and Efficiency and Performance for Connectivity Constrained Computing (EP3C).

Research Focus

The AI Hardware research program is sub-divided into six major categories:

• Novel Architectures for Accelerating AI Computation

• Modeling and Simulation/Emulation of AI Hardware for Early System Exploration

• Power Efficient AI Hardware System Design

• HW/SW Co-design of AI Compute Systems

• Fairness, Robustness, Privacy, and Explainability of Models and Algorithms for AI Hardware

• Interplay of AI and System Architecture/Microarchitecture Design

In each category, there may be research covering large systems to small (datacenter and the edge/end node) as well as a broad range of applications, including high-performance processors for data centers, automotive, industrial, mobile computing and communication, and healthcare.

AIHW Metrics

  1. Current

    28 Research Tasks
    23 Universities
    63 Students
    32 Faculty Researchers
    88 Liaison Personnel
  2. This Year

    17 Task Starts
    81 Research Publications
  3. Last Year

    3 Task Starts
    70 Research Publications
  4. Since Inception

    32 Research Tasks
    27 Universities
    72 Students
    39 Faculty Researchers
    102 Liaison Personnel
    220 Research Publications
Updated: 7-Aug-2020, 12:05 a.m. ET

4819 Emperor Blvd, Suite 300 Durham, NC 27703 Voice: (919) 941-9400 Fax: (919) 941-9450

Important Information for the SRC website. This site uses cookies to store information on your computer. By continuing to use our site, you consent to our cookies. If you are not happy with the use of these cookies, please review our Cookie Policy to learn how they can be disabled. By disabling cookies, some features of the site will not work.