Measuring Equality in Machine Learning Security Defenses: A Case Study in Speech Recognition

• URL: http://arxiv.org/abs/2302.08973v6
• Date: Wed, 23 Aug 2023 01:05:39 GMT
• Title: Measuring Equality in Machine Learning Security Defenses: A Case Study in Speech Recognition
• Authors: Luke E. Richards, Edward Raff, Cynthia Matuszek
• Abstract summary: This work considers approaches to defending learned systems and how security defenses result in performance inequities across different sub-populations. We find that many methods that have been proposed can cause direct harm, like false rejection and unequal benefits from robustness training. We present a comparison of equality between two rejection-based defenses: randomized smoothing and neural rejection, finding randomized smoothing more equitable due to the sampling mechanism for minority groups.
• Score: 56.69875958980474
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Over the past decade, the machine learning security community has developed a myriad of defenses for evasion attacks. An understudied question in that community is: for whom do these defenses defend? This work considers common approaches to defending learned systems and how security defenses result in performance inequities across different sub-populations. We outline appropriate parity metrics for analysis and begin to answer this question through empirical results of the fairness implications of machine learning security methods. We find that many methods that have been proposed can cause direct harm, like false rejection and unequal benefits from robustness training. The framework we propose for measuring defense equality can be applied to robustly trained models, preprocessing-based defenses, and rejection methods. We identify a set of datasets with a user-centered application and a reasonable computational cost suitable for case studies in measuring the equality of defenses. In our case study of speech command recognition, we show how such adversarial training and augmentation have non-equal but complex protections for social subgroups across gender, accent, and age in relation to user coverage. We present a comparison of equality between two rejection-based defenses: randomized smoothing and neural rejection, finding randomized smoothing more equitable due to the sampling mechanism for minority groups. This represents the first work examining the disparity in the adversarial robustness in the speech domain and the fairness evaluation of rejection-based defenses.

Related papers

• Among Us: Adversarially Robust Collaborative Perception by Consensus [50.73128191202585]
  Multiple robots could perceive a scene (e.g., detect objects) collaboratively better than individuals. We propose ROBOSAC, a novel sampling-based defense strategy generalizable to unseen attackers. We validate our method on the task of collaborative 3D object detection in autonomous driving scenarios.
  arXiv  Detail & Related papers  (2023-03-16T17:15:25Z)
• Adversarial Purification with the Manifold Hypothesis [14.085013765853226]
  We develop an adversarial purification method with this framework. Our approach can provide adversarial robustness even if attackers are aware of the existence of the defense.
  arXiv  Detail & Related papers  (2022-10-26T01:00:57Z)
• Towards Fair Classification against Poisoning Attacks [52.57443558122475]
  We study the poisoning scenario where the attacker can insert a small fraction of samples into training data. We propose a general and theoretically guaranteed framework which accommodates traditional defense methods to fair classification against poisoning attacks.
  arXiv  Detail & Related papers  (2022-10-18T00:49:58Z)
• Resisting Adversarial Attacks in Deep Neural Networks using Diverse Decision Boundaries [12.312877365123267]
  Deep learning systems are vulnerable to crafted adversarial examples, which may be imperceptible to the human eye, but can lead the model to misclassify. We develop a new ensemble-based solution that constructs defender models with diverse decision boundaries with respect to the original model. We present extensive experimentations using standard image classification datasets, namely MNIST, CIFAR-10 and CIFAR-100 against state-of-the-art adversarial attacks.
  arXiv  Detail & Related papers  (2022-08-18T08:19:26Z)
• Searching for an Effective Defender: Benchmarking Defense against Adversarial Word Substitution [83.84968082791444]
  Deep neural networks are vulnerable to intentionally crafted adversarial examples. Various methods have been proposed to defend against adversarial word-substitution attacks for neural NLP models.
  arXiv  Detail & Related papers  (2021-08-29T08:11:36Z)
• Improving the Adversarial Robustness for Speaker Verification by Self-Supervised Learning [95.60856995067083]
  This work is among the first to perform adversarial defense for ASV without knowing the specific attack algorithms. We propose to perform adversarial defense from two perspectives: 1) adversarial perturbation purification and 2) adversarial perturbation detection. Experimental results show that our detection module effectively shields the ASV by detecting adversarial samples with an accuracy of around 80%.
  arXiv  Detail & Related papers  (2021-06-01T07:10:54Z)
• Adversarial robustness via stochastic regularization of neural activation sensitivity [24.02105949163359]
  We suggest a novel defense mechanism that simultaneously addresses both defense goals. We flatten the gradients of the loss surface, making adversarial examples harder to find. In addition, we push the decision away from correctly classified inputs by leveraging Jacobian regularization.
  arXiv  Detail & Related papers  (2020-09-23T19:31:55Z)
• Defensive Few-shot Learning [77.82113573388133]
  This paper investigates a new challenging problem called defensive few-shot learning. It aims to learn a robust few-shot model against adversarial attacks. The proposed framework can effectively make the existing few-shot models robust against adversarial attacks.
  arXiv  Detail & Related papers  (2019-11-16T05:57:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.