• Skip to main content
  • Skip to secondary menu
  • Skip to primary sidebar
  • Skip to footer
Montreal AI Ethics Institute

Montreal AI Ethics Institute

Democratizing AI ethics literacy

  • Articles
    • Public Policy
    • Privacy & Security
    • Human Rights
      • Ethics
      • JEDI (Justice, Equity, Diversity, Inclusion
    • Climate
    • Design
      • Emerging Technology
    • Application & Adoption
      • Health
      • Education
      • Government
        • Military
        • Public Works
      • Labour
    • Arts & Culture
      • Film & TV
      • Music
      • Pop Culture
      • Digital Art
  • Columns
    • AI Policy Corner
    • Recess
    • Tech Futures
  • The AI Ethics Brief
  • AI Literacy
    • Research Summaries
    • AI Ethics Living Dictionary
    • Learning Community
  • The State of AI Ethics Report
    • Volume 8 (November 2026)
    • Volume 7 (November 2025)
    • Volume 6 (February 2022)
    • Volume 5 (July 2021)
    • Volume 4 (April 2021)
    • Volume 3 (Jan 2021)
    • Volume 2 (Oct 2020)
    • Volume 1 (June 2020)
  • About
    • Our Contributions Policy
    • Our Open Access Policy
    • Contact
    • Donate

Research Summary: Explaining and Harnessing Adversarial Examples

June 28, 2020

Summary contributed by Shannon Egan, Research Fellow at Building 21 and pursuing a master’s in physics at UBC.

*Author & link to original paper at the bottom.


Click here for the FULL summary in PDF form

(Short-form summary below)

A bemusing weakness of many supervised machine learning (ML) models, including neural networks (NNs), are adversarial examples (AEs).  AEs are inputs generated by adding a small perturbation to a correctly-classified input, causing the model to misclassify the resulting AE with high confidence.  Goodfellow et al. propose a linear explanation of AEs, in which the vulnerability of ML models to AEs is considered a by-product of their linear behaviour and high-dimensional feature space.  In other words, small perturbations on an input can alter its classification because the change in NN activation (as result of the perturbation) scales with the size of the input vector.

Identifying ways to effectively handle AEs is of interest for problems like image classification, where the input consists of intensity data for many thousands of pixels.  A method of generating AEs called “fast gradient sign method” badly fools a maxout network, leading to a 89.4% error rate on a perturbed MNIST test set.  The authors propose an “adversarial training” scheme for NNs, in which an adversarial term is added to the loss function during training. 

This dramatically improves the error rate of the same maxout network to 17.4% on AEs generated by the fast gradient sign method. The linear interpretation of adversarial examples suggests an approach to adversarial training which improves a model’s ability to classify AEs, and helps interpret properties of AE classification which the previously proposed nonlinearity and overfitting hypotheses do not explain. 


Click here for the full summary in PDF form.

Original paper by Ian J. Goodfellow, Jonathan Shlens and Christian Szegedy: https://arxiv.org/abs/1412.6572

Want quick summaries of the latest research & reporting in AI ethics delivered to your inbox? Subscribe to the AI Ethics Brief. We publish bi-weekly.

Primary Sidebar

🔍 SEARCH

Spotlight

Vertically- and horizontally-placed chess boards and chess pieces

Tech Futures: At the Frontier of Fear, Uncertainty and Doubt

Tech Futures: Introducing the Resist List

An abstract spiral of dark circles appears at the centre, resembling a tornado. Several vintage magazine covers and advertisements are being drawn toward the spiral. The artworks that have already been pulled into it are becoming distorted and replaced with clusters of numbers representing their numerical embeddings.

Tech Futures: Better Imagination for Better Tech Futures

This image is a collage with a colourful Japanese vintage landscape showing a mountain, hills, flowers and other plants and a small stream. There are 3 large black data servers placed in the bottom half of the image, with a cloud of black smoke emitting from them, partly obscuring the scenery.

Tech Futures: Crafting Participatory Tech Futures

A network diagram with lots of little emojis, organised in clusters.

Tech Futures: AI For and Against Knowledge

related posts

    • The Evolution of War: How AI has Changed Military Weaponry and Technology

      The Evolution of War: How AI has Changed Military Weaponry and Technology

    • Computers, Creativity and Copyright: Autonomous Robot’s Status, Authorship, and Outdated Copyright L...

      Computers, Creativity and Copyright: Autonomous Robot’s Status, Authorship, and Outdated Copyright L...

    • On Measuring Fairness in Generative Modelling (NeurIPS 2023)

      On Measuring Fairness in Generative Modelling (NeurIPS 2023)

    • Defining a Research Testbed for Manned-Unmanned Teaming Research

      Defining a Research Testbed for Manned-Unmanned Teaming Research

    • Responsible sourcing and the professionalization of data work

      Responsible sourcing and the professionalization of data work

    • Mapping AI Arguments in Journalism and Communication Studies

      Mapping AI Arguments in Journalism and Communication Studies

    • AI Ethics Maturity Model

      AI Ethics Maturity Model

    • Automated Interviewer or Augmented Survey? Collecting Social Data with Large Language Models

      Automated Interviewer or Augmented Survey? Collecting Social Data with Large Language Models

    • To Be or Not to Be Algorithm Aware: A Question of a New Digital Divide? (Research Summary)

      To Be or Not to Be Algorithm Aware: A Question of a New Digital Divide? (Research Summary)

    • Putting AI ethics to work: are the tools fit for purpose?

      Putting AI ethics to work: are the tools fit for purpose?

Partners

  •  
    U.S. Artificial Intelligence Safety Institute Consortium (AISIC) at NIST

  • Partnership on AI

  • The LF AI & Data Foundation

  • The AI Alliance

Footer


Articles

Columns

AI Literacy

The State of AI Ethics Report


 

About Us


Founded in 2018, the Montreal AI Ethics Institute (MAIEI) is an international non-profit organization equipping citizens concerned about artificial intelligence and its impact on society to take action.

Contact

Donate


  • © 2025 MONTREAL AI ETHICS INSTITUTE.
  • This work is licensed under a Creative Commons Attribution 4.0 International License.
  • Learn more about our open access policy here.
  • Creative Commons License

    Save hours of work and stay on top of Responsible AI research and reporting with our bi-weekly email newsletter.