PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 23, 2026ACM Computing Surveys1 citationsOpen Access

TinyML Security: Attacks and Defenses in Resource-Constrained Machine Learning

TinyML Security: Attacks, Defenses, and Open Challenges in Resource-Constrained Machine Learning Systems

View Full Paper
Ask AI
Bookmark
Share

Authors

JHJacob HuckelberryYZYuke ZhangASAnna Sansone

Discussion

Loading...

Member takes

Overview

Systematic review reveals security vulnerabilities in TinyML systems, indicating the need for optimized countermeasures.

Key Points

  • The aim is to survey the security landscape of TinyML and identify unique vulnerabilities.
  • Systematic literature review of TinyML publications
  • Development of a resource-based taxonomy for IoT, EdgeML, and TinyML
  • Analysis of attack classes across hardware, software, and model layers
  • Assessment of threat severity using the Common Vulnerability Scoring System
  • Evaluation of traditional countermeasures against TinyML constraints
  • Fewer than 5% of publications address security in TinyML
  • Identification of critical gaps in current defenses
  • Need for lightweight security solutions due to prohibitive overhead
  • Classification of eleven attack classes and their impact on TinyML security

Cite This Study

Huckelberry et al. (2026) studied this question.

synapsesocial.com/papers/699ba0a772792ae9fd870963https://doi.org/10.1145/3793197
View Full Paper
Ask AI
Bookmark
Share