Hardware support for trustworthy machine learning: a survey

Md Shohidul Islam, Ihsen Alouani, Khaled N. Khasawneh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Downloads (Pure)

Abstract

Machine Learning (ML) are used in an increasing number of applications as they continue to deliver state-of-the-art performance across many areas including computer vision natural language processing (NLP), robotics, autonomous driving, and healthcare. While rapid progress in all aspects of ML development and deployment is occurring, there is a rising concern about the trustworthiness of these models, especially from security and privacy perspectives. Several attacks that jeopardize ML models’ integrity (e.g. adversarial attacks) and confidentiality (e.g. membership inference attacks) have been investigated in the literature. This, in return, triggered substantial work to protect ML models and advance their trustworthiness. Defenses generally act on the input data, the objective function, or the network structure to mitigate adversarial effects. However, these proposed defenses require substantial changes to the architecture, retraining procedure, or incorporate additional input data processing overheads. In addition, often these proposed defenses require high power and computational requirements, which make them challenging to deploy in embedded systems and Edge devices. Towards addressing the need for robust ML at acceptable overheads, recent works have investigated hardware-emanated solutions to enhance ML security and privacy. In this paper, we summarize recent works in the area of hardware support for trustworthy ML. In addition, we provide guidelines for future research in the area by identifying open problems that need to be addressed.
Original languageEnglish
Title of host publication2024 25th International Symposium on Quality Electronic Design (ISQED): proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages6
ISBN (Electronic)9798350309270
ISBN (Print)9798350309287
DOIs
Publication statusPublished - 16 May 2024
Event2024 25th International Symposium on Quality Electronic Design (ISQED) - San Francisco, United States
Duration: 03 Apr 202405 Apr 2024

Publication series

NameInternational Symposium on Quality Electronic Design (ISQED): proceedings
ISSN (Print)1948-3287
ISSN (Electronic)1948-3295

Conference

Conference2024 25th International Symposium on Quality Electronic Design (ISQED)
Country/TerritoryUnited States
CitySan Francisco
Period03/04/202405/04/2024

Publications and Copyright Policy

This work is licensed under Queen’s Research Publications and Copyright Policy.

Keywords

  • Surveys
  • Privacy
  • Computational modeling
  • Machine learning
  • Medical services
  • Linear programming
  • Natural language processing

Fingerprint

Dive into the research topics of 'Hardware support for trustworthy machine learning: a survey'. Together they form a unique fingerprint.

Cite this