mailitics

Tag: adversarial

  • On damage of interpolation to adversarial robustness in regression

    On damage of interpolation to adversarial robustness in regression arXiv:2601.16070v1 Announce Type: new Abstract: Deep neural networks (DNNs) typically involve a large number of parameters and are trained to achieve zero or near-zero training error. Despite such interpolation, they often exhibit strong generalization performance on unseen data, a phenomenon that has motivated extensive theoretical investigations.…

  • Hellinger loss function for Generative Adversarial Networks

    Hellinger loss function for Generative Adversarial Networks arXiv:2512.12267v1 Announce Type: new Abstract: We propose Hellinger-type loss functions for training Generative Adversarial Networks (GANs), motivated by the boundedness, symmetry, and robustness properties of the Hellinger distance. We define an adversarial objective based on this divergence and study its statistical properties within a general parametric framework. We…

  • Impact of Positional Encoding: Clean and Adversarial Rademacher Complexity for Transformers under In-Context Regression

    Impact of Positional Encoding: Clean and Adversarial Rademacher Complexity for Transformers under In-Context Regression arXiv:2512.09275v1 Announce Type: new Abstract: Positional encoding (PE) is a core architectural component of Transformers, yet its impact on the Transformer’s generalization and robustness remains unclear. In this work, we provide the first generalization analysis for a single-layer Transformer under in-context…

  • Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization

    Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization arXiv:2510.20883v1 Announce Type: new Abstract: Adversarial training has emerged as a key technique to enhance model robustness against adversarial input perturbations. Many of the existing methods rely on computationally expensive min-max problems that limit their application in practice. We propose a novel formulation of adversarial…

  • A unified Bayesian framework for adversarial robustness

    A unified Bayesian framework for adversarial robustness arXiv:2510.09288v1 Announce Type: new Abstract: The vulnerability of machine learning models to adversarial attacks remains a critical security challenge. Traditional defenses, such as adversarial training, typically robustify models by minimizing a worst-case loss. However, these deterministic approaches do not account for uncertainty in the adversary’s attack. While stochastic…

  • On the Adversarial Robustness of Learning-based Conformal Novelty Detection

    On the Adversarial Robustness of Learning-based Conformal Novelty Detection arXiv:2510.00463v1 Announce Type: new Abstract: This paper studies the adversarial robustness of conformal novelty detection. In particular, we focus on AdaDetect, a powerful learning-based framework for novelty detection with finite-sample false discovery rate (FDR) control. While AdaDetect provides rigorous statistical guarantees under benign conditions, its behavior…

  • On the existence of consistent adversarial attacks in high-dimensional linear classification

    On the existence of consistent adversarial attacks in high-dimensional linear classification arXiv:2506.12454v1 Announce Type: new Abstract: What fundamentally distinguishes an adversarial attack from a misclassification due to limited model expressivity or finite data? In this work, we investigate this question in the setting of high-dimensional binary classification, where statistical effects due to limited data availability…

  • Robust Learnability of Sample-Compressible Distributions under Noisy or Adversarial Perturbations

    Robust Learnability of Sample-Compressible Distributions under Noisy or Adversarial Perturbations arXiv:2506.06613v1 Announce Type: new Abstract: Learning distribution families over $mathbb{R}^d$ is a fundamental problem in unsupervised learning and statistics. A central question in this setting is whether a given family of distributions possesses sufficient structure to be (at least) information-theoretically learnable and, if so, to…

  • Generative Adversarial Networks for High-Dimensional Item Factor Analysis: A Deep Adversarial Learning Algorithm

    Generative Adversarial Networks for High-Dimensional Item Factor Analysis: A Deep Adversarial Learning Algorithm arXiv:2502.10650v1 Announce Type: new Abstract: Advances in deep learning and representation learning have transformed item factor analysis (IFA) in the item response theory (IRT) literature by enabling more efficient and accurate parameter estimation. Variational Autoencoders (VAEs) have been one of the most…

  • Provably Safeguarding a Classifier from OOD and Adversarial Samples: an Extreme Value Theory Approach

    Provably Safeguarding a Classifier from OOD and Adversarial Samples: an Extreme Value Theory Approach arXiv:2501.10202v1 Announce Type: new Abstract: This paper introduces a novel method, Sample-efficient Probabilistic Detection using Extreme Value Theory (SPADE), which transforms a classifier into an abstaining classifier, offering provable protection against out-of-distribution and adversarial samples. The approach is based on a…