Tag
This paper analyzes the transferability of adversarial attacks in federated learning systems and proposes a defense mechanism based on adversarial training to enhance model robustness.
USAD proposes two new statistics, Variance Discrepancy and Perturbation-based Covariance Discrepancy, to capture global and local uncertainty patterns of adversarial examples, achieving superior detection performance over baseline methods.
This paper presents a certification-inspired mechanism for automatic speech recognition that uses a dual-gate diagnostic pipeline (Two-Sided Atomic Audit and Rank-Based Tournament) to provide certified robustness and achieve up to a 55% relative reduction in word error rate across diverse architectures.
This paper introduces the apothem measure for computing trustworthy robustness certifications in neural networks, proves intractability of volume-optimal certifications, and presents the ParallelepipedoNN system achieving twofold improvement in minimum edge length on MNIST and Fashion MNIST.
Abliteration launches a made-to-order synthetic training data workflow that generates negative, rare, and adversarial examples for classifiers, with schema, real-world facts, labels, provenance, and export to platforms like Hugging Face.
Researchers study how adversarial robustness transfers across different perturbation types in deep neural networks, evaluating 32 attacks of 5 types on ImageNet models. Results show that robustness to one perturbation type doesn't always transfer to others and may sometimes hurt robustness elsewhere.
OpenAI introduces Activation Atlases, a technique for visualizing and understanding the internal representations of neural networks, enabling humans to discover spurious correlations and unexpected behaviors such as fooling image classifiers by adding noodles to images.
Researchers demonstrated adversarial images that reliably fool neural network classifiers across multiple scales and perspectives, challenging assumptions about the robustness of multi-scale image capture systems used in autonomous vehicles.
This article examines adversarial attacks on machine learning models and demonstrates why gradient masking—a defensive technique that attempts to deny attackers access to useful gradients—is fundamentally ineffective. The paper shows that attackers can circumvent gradient masking by training substitute models that mimic the defended model's behavior, making the defense strategy ultimately futile.
AI systems must incorporate probabilistic reasoning to handle uncertainty for robust and trustworthy decision-making, as discussed by Zoubin Ghahramani regarding the mathematics of uncertainty in intelligent systems.