A Sound Mind in a Vulnerable Body - Practical Hardware Attacks on Deep Learning

A Sound Mind in a Vulnerable Body - Practical Hardware Attacks on Deep Learning

USENIX Enigma Conference via YouTube Direct link

Intro

1 of 14

1 of 14

Intro

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

A Sound Mind in a Vulnerable Body - Practical Hardware Attacks on Deep Learning

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Recent Work on Secure Machine Learning
  3. 3 Conventional View on ML Models' Robustness
  4. 4 We Propose A New Perspective!
  5. 5 Hardware Attacks Can Break Mathematically-Proven Guarantees
  6. 6 (Weak) Hardware Attacks Can Be Exploited in the Cloud
  7. 7 Prior Work's Perspective on a Model's Robustness
  8. 8 The Worst-Case Perturbation
  9. 9 Threat Model - Single-Bit Adversaries
  10. 10 Evaluate the Weakest Attacker with Multiple Bit-flips
  11. 11 Our Attack: Reconstruction of DNN Architectures from the Trace
  12. 12 We Can Identify the Layers Accessed While Computing
  13. 13 Solution: Generate All Candidate Architectures
  14. 14 Solution: Eliminate incompatible Candidates

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.