arXiv research

KSGAN uses KS distance for deep generative modeling.

Mathematical conditions and practical computations for adversarial robustness measures are established.

Paper proposes methods to estimate minimal adversarial perturbations for deep neural networks.

In this article, we introduce a new mode for training Generative Adversarial Networks (GANs). Rather than minimizing the distance of evidence distribution $\tilde{p}(x)$ and the generative distribution $q(x)$, we minimize the distance of $\tilde{p}(x_r)q(x_f)$ and $\tilde{p}(x_f)q(x_r)$. This adversarial pattern can be…

Paper uses Sinkhorn distances to improve imitation learning effectiveness.

We present Optimal Transport GAN (OT-GAN), a variant of generative adversarial nets minimizing a new metric measuring the distance between the generator distribution and the data distribution. This metric, which we call mini-batch energy distance, combines optimal transport in primal form with an energy distance define…

Differentiable adversarial attacks improve model robustness in MTPP models.

Tricks adversarial attacks to target specific classes, improving classifier accuracy.

Study robust distribution estimation with Wasserstein distance, achieving optimal risk.

Study improves adversarial classification using distributionally robust models.

Novel neural network approximates exact distance for robust classification.

New geometric insights reveal properties of adversarial training problems.

Advances robustness of metric learning by adversarial margin in input space.

Adversarial attacks for image classification are small perturbations to images that are designed to cause misclassification by a model. Adversarial attacks formally correspond to an optimization problem: find a minimum norm image perturbation, constrained to cause misclassification. A number of effective attacks have b…

We propose a new active learning strategy designed for deep neural networks. The goal is to minimize the number of data annotation queried from an oracle during training. Previous active learning strategies scalable for deep networks were mostly based on uncertain sample selection. In this work, we focus on examples ly…

Neural networks have been proven to be vulnerable to a variety of adversarial attacks. From a safety perspective, highly sparse adversarial attacks are particularly dangerous. On the other hand the pixelwise perturbations of sparse attacks are typically large and thus can be potentially detected. We propose a new black…

Paper defends sensitive attributes in GNNs from inference attacks.

Paper proposes a new black-box attack approach to minimize visual distortion.

Generative neural network models, including Generative Adversarial Network (GAN) and Auto-Encoders (AE), are among the most popular neural network models to generate adversarial data. The GAN model is composed of a generator that produces synthetic data and of a discriminator that discriminates between the generator's …

We study adversarial robustness of neural networks from a margin maximization perspective, where margins are defined as the distances from inputs to a classifier's decision boundary. Our study shows that maximizing margins can be achieved by minimizing the adversarial loss on the decision boundary at the "shortest succ…

Paper proposes a new method for learning compact representations of sequential data.

WDAIL uses Wasserstein distance for more effective reward shaping in IL.

The study improves PAC-Bayesian bounds for adversarial generative models.

New method defends against unseen threat models using perceptual adversarial training.

Efficiently learns distributions corrupted by both global and local adversarial modifications.

A new method for fair classification using characteristic function distance.

WAPPO optimizes feature distributions for better visual transfer in RL.

Generative adversarial networks (GANs) have received a tremendous amount of attention in the past few years, and have inspired applications addressing a wide range of problems. Despite its great potential, GANs are difficult to train. Recently, a series of papers (Arjovsky & Bottou, 2017a; Arjovsky et al. 2017b; and Gu…

Develops a new non-adversarial framework for better generative models.

In the last couple of years, several adversarial attack methods based on different threat models have been proposed for the image classification problem. Most existing defenses consider additive threat models in which sample perturbations have bounded L_p norms. These defenses, however, can be vulnerable against advers…

DIP-FAT improves adversarial training by diversifying perturbations.

This paper presents an end-to-end text-independent speaker verification framework by jointly considering the speaker embedding (SE) network and automatic speech recognition (ASR) network. The SE network learns to output an embedding vector which distinguishes the speaker characteristics of the input utterance, while th…

AAS optimizes neural network PDE approximations by adaptively sampling.

Despite numerous attempts to defend deep learning based image classifiers, they remain susceptible to the adversarial attacks. This paper proposes a technique to identify susceptible classes, those classes that are more easily subverted. To identify the susceptible classes we use distance-based measures and apply them …

Recent studies show that widely used deep neural networks (DNNs) are vulnerable to carefully crafted adversarial examples. Many advanced algorithms have been proposed to generate adversarial examples by leveraging the $\mathcal{L}_p$ distance for penalizing perturbations. Researchers have explored different defense met…

Proposes ATM method to improve domain adaptation.

RS-Del provides robustness for sequence classifiers against edit distance attacks.

Adversarial consistency depends on the uniqueness of adversarial Bayes classifiers.

WGANs use optimal 1-Wasserstein distance to generate distributions.

Most random ReLU networks are vulnerable to small, Euclidean adversarial perturbations.

Proposes Topology Distance for evaluating GANs.

Simple algorithm achieves distance to calibration error of at most 2√T+1.

It has been shown that neural network classifiers are not robust. This raises concerns about their usage in safety-critical systems. We propose in this paper a regularization scheme for ReLU networks which provably improves the robustness of the classifier by maximizing the linear regions of the classifier as well as t…

Minimum expected distance estimation (MEDE) algorithms have been widely used for probabilistic models with intractable likelihood functions and they have become increasingly popular due to their use in implicit generative modeling (e.g. Wasserstein generative adversarial networks, Wasserstein autoencoders). Emerging fr…

Validates neural networks inputs to protect against adversarial examples.

Study adversarial classification with data corruption up to ε, deriving geometric flows.

ClusTR improves clustering-based models' robustness without adversarial training.

Domain adaptation aims at generalizing a high-performance learner on a target domain via utilizing the knowledge distilled from a source domain which has a different but related data distribution. One solution to domain adaptation is to learn domain invariant feature representations while the learned representations sh…