Discrete Randomized Smoothing Meets Quantum Computing

Tom Wollschlager, Aman Saxena, Nicola Franco, Jeanette Miriam Lorenz, Stephan Gunnemann

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Breakthroughs in machine learning (ML) and advances in quantum computing (QC) drive the interdisciplinary field of quantum machine learning to new levels. However, due to the susceptibility of ML models to adversarial attacks, practical use raises safety-critical concerns. Existing Randomized Smoothing (RS) certification methods for classical machine learning models are computationally intensive. In this paper, we propose the combination of QC and the concept of discrete randomized smoothing to speed up the stochastic certification of ML models for discrete data. We show how to encode all the perturbations of the input binary data in superposition and use Quantum Amplitude Estimation (QAE) to obtain a quadratic reduction in the number of calls to the model that are required compared to traditional randomized smoothing techniques. In addition, we propose a new binary threat model to allow for an extensive evaluation of our approach on images, graphs, and text.

Original languageEnglish
Title of host publicationTechnical Papers Program
EditorsCandace Culhane, Greg T. Byrd, Hausi Muller, Yuri Alexeev, Yuri Alexeev, Sarah Sheldon
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1535-1546
Number of pages12
ISBN (Electronic)9798331541378
DOIs
StatePublished - 2024
Event5th IEEE International Conference on Quantum Computing and Engineering, QCE 2024 - Montreal, Canada
Duration: 15 Sep 202420 Sep 2024

Publication series

NameProceedings - IEEE Quantum Week 2024, QCE 2024
Volume1

Conference

Conference5th IEEE International Conference on Quantum Computing and Engineering, QCE 2024
Country/TerritoryCanada
CityMontreal
Period15/09/2420/09/24

Keywords

  • Certifiable Robustness
  • Quantum Amplitude Estimation
  • Quantum Machine Learning
  • Randomized Smoothing

Fingerprint

Dive into the research topics of 'Discrete Randomized Smoothing Meets Quantum Computing'. Together they form a unique fingerprint.

Cite this