arXiv research

New method for analyzing compositional data, addressing biases in summary statistics.

SAMS-VAE models cellular perturbations using sparse additive mechanisms.

New method estimates heterogeneous treatment effects with improved guarantees.

Proposes a new framework for balancing average- and worst-case performance in machine learning.

Lower class selectivity makes networks more robust to natural perturbations but more vulnerable to adversarial attacks.

We consider in this work a system of two stochastic differential equations named the perturbed compositional gradient flow. By introducing a separation of fast and slow scales of the two equations, we show that the limit of the slow motion is given by an averaged ordinary differential equation. We then demonstrate that…

This paper presents a new approach, called perturb-max, for high-dimensional statistical inference that is based on applying random perturbations followed by optimization. This framework injects randomness to maximum a-posteriori (MAP) predictors by randomly perturbing the potential function for the input. A classic re…

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…

Simple regional perturbations maintain model transferability while reducing adversarial example distortion.

Study privacy and accuracy in high-dimensional LASSO with perturbation mechanisms.

Nowadays, machine learning based Automatic Speech Recognition (ASR) technique has widely spread in smartphones, home devices, and public facilities. As convenient as this technology can be, a considerable security issue also raises -- the users' speech content might be exposed to malicious ASR monitoring and cause seve…

We provide adaptive inference methods, based on $\ell_1$ regularization, for regular (semi-parametric) and non-regular (nonparametric) linear functionals of the conditional expectation function. Examples of regular functionals include average treatment effects, policy effects, and derivatives. Examples of non-regular f…

Paper introduces EnCounteR for estimating causal effects using encouragement data.

PEP improves deep network performance and calibration by perturbing optimal parameters.

This research investigates reliable local explanations for machine listening models.

New metric scores perturbations across populations, not cells, improving model comparison.

New optimization method improves generalization across various tasks.

Generative Intervention Models predict perturbation effects without knowing the underlying mechanisms.

This project improves model robustness to affine transformations.

The paper tackles extrapolation of gene knockouts effects on RNA counts.

We present a geometric proof of the averaging theorem for perturbed dynamical systems on a Riemannian manifold, in the case where the flow of the unperturbed vector field is periodic and the $\mathbb{S}^{1}$-action associated to this vector field is not necessarily trivial. We generalize the averaging procedure \cite{A…

Adversarial perturbations fool deepfake detectors with high accuracy.

The universal perturbative invariants of rational homology spheres can be extracted from the Chern-Simons partition function by combining perturbative and nonperturbative results. We spell out the general procedure to compute these invariants, and we work out in detail the case of Seifert spaces. By extending some prev…

Study reveals class-dependent effects in perturbation-based feature attribution metrics for time series classification.

This paper explores adversarial training limits and improves model robustness against norm-bounded perturbations.

LMC improves sampling from complex distributions using quasi-random sequences.

Deterministic method for certifying neural network robustness.

New method $μP^2$ improves neural network training by scaling perturbations layerwise.

Gradient perturbation, widely used for differentially private optimization, injects noise at every iterative update to guarantee differential privacy. Previous work first determines the noise level that can satisfy the privacy requirement and then analyzes the utility of noisy gradient updates as in the non-private cas…

Convolutional neural networks (CNN) have become one of the most popular machine learning tools and are being applied in various tasks, however, CNN models are vulnerable to universal perturbations, which are usually human-imperceptible but can cause natural images to be misclassified with high probability. One of the s…

GNNGuard defends Graph Neural Networks against structural perturbations.

Recent research has revealed that the output of Deep Neural Networks (DNN) can be easily altered by adding relatively small perturbations to the input vector. In this paper, we analyze an attack in an extremely limited scenario where only one pixel can be modified. For that we propose a novel method for generating one-…

Robustness is an increasingly important property of machine learning models as they become more and more prevalent. We propose a defense against adversarial examples based on a k-nearest neighbor (kNN) on the intermediate activation of neural networks. Our scheme surpasses state-of-the-art defenses on MNIST and CIFAR-1…

PAC-Bayesian bounds estimate adversarial robustness.

Study develops method for estimating causal effects in continuous variables.

WassersteinGrad improves weather forecasting explanations by addressing geometric misalignment issues.

Item recommendation is a personalized ranking task. To this end, many recommender systems optimize models with pairwise ranking objectives, such as the Bayesian Personalized Ranking (BPR). Using matrix Factorization (MF) --- the most widely used model in recommendation --- as a demonstration, we show that optimizing it…

Improved generalization with iterate averaging and adaptive algorithms.

Bayesian model averaging improves causal effect estimation by averaging over multiple models.

We propose an online algorithm for cumulative regret minimization in a stochastic multi-armed bandit. The algorithm adds $O(t)$ i.i.d. pseudo-rewards to its history in round $t$ and then pulls the arm with the highest average reward in its perturbed history. Therefore, we call it perturbed-history exploration (PHE). Th…

Optimal sample complexity analysis for plug-in approach in average-reward MDPs.

DBPA assesses LLM perturbations using frequentist hypothesis testing.

This paper explores how facial recognition systems can be fooled by adversarial attacks.

This dissertation advances the theoretical foundation of local optimization methods in Federated Learning.

Perturbative string amplitudes are correctly derived from the string geometry theory, which is one of the candidates of a non-perturbative formulation of string theory. In order to derive non-perturbative effects rather easily, we formulate topological string geometry theory. We derive the perturbative partition functi…

Adversarial weight perturbations can inject backdoors into trained neural models.

The problem of attribution is concerned with identifying the parts of an input that are responsible for a model's output. An important family of attribution methods is based on measuring the effect of perturbations applied to the input. In this paper, we discuss some of the shortcomings of existing approaches to pertur…

Wide networks learn from adversarial perturbations effectively.