Search Results

The reliability of deep neural networks is critical for industrial applications as well as human safety and security. However, artificial deep neural networks have been found vulnerable to multiple kinds of natural, artificial, and adversarial ...

This paper introduces a novel task to assess the faithfulness of large language models (LLMs) using local perturbations and self-explanations. Many LLMs often require additional context to answer certain questions correctly. For this purpose, we propose a ...

This article introduces a novel approach for building a robust Automatic Gesture Recognition system based on Surface Electromyographic (sEMG) signals, acquired at the forearm level. Our main contribution is to propose new constrained learning strategies ...

In the past few years, it has become increasingly evident that deep neural networks are not resilient enough to withstand adversarial perturbations in input data, leaving them vulnerable to attack. Various authors have proposed strong adversarial attacks ...

White‐box cryptography protects the secret key extraction from software implementations of cryptographic primitives. The principle is to white‐box the key by using obfuscation and hide the key in the whole encryption process to ensure that the ...

Gradient descent (GD) and stochastic gradient descent (SGD) are the workhorses of large-scale machine learning. While classical theory focused on analyzing the performance of these methods in convex optimization problems, the most notable successes in ...

Organic semiconductors are intrinsic randomly inhomogeneous materials where charge transport occurs by hopping of the carriers between localized sites having a distribution of energy levels. However, the average carrier density seems to be accurately ...

In many realistic applications, the collected inputs of DNN face a big challenge: perturbations. Although the perturbations are imperceptible, they may cause incorrect prediction results. This paper proposes prediction stability to quantitatively ...

In this paper, we aim to explain the decisions of neural networks by utilizing multimodal information. That is counter-intuitive attributes and counter visual examples which appear when perturbed samples are introduced. Different from previous work on ...

Given dynamics and constraints, the viability kernel gathers points from which there exists an evolution which remains in the constraint set. In this paper, we aim at providing more information: we introduce and study the robustness map which associates ...

In this paper, nonlinear control design scheme for a class of nonlinear systems is proposed based on operator coprime factorization theory. In detail, two stable controllers are provided to design a Bezout identity by combining left factorization (...

With the proliferation of the touchscreen technology, interactive displays are becoming an integrated part of the modern vehicle environment. However, due to road and driving conditions, the user input on such displays can be perturbed resulting in ...

Aeronautics and air transport is a vital sector of our society and economy. Air transport logistics is one of the key players to support efficient globalization; however, sustainable mobility is at stake, due to facts such as the interdependencies with ...

In this paper we describe the design of a multiagent system based on simple interaction rules that can generate different overall behaviours, from asymptotically stable to chaotic, verified by the corresponding largest Lyapunov exponent. We show that ...

We consider parameter dependent semilinear evolution problems for which, at the limit value of the parameter, the problem is finite dimensional. We introduce an abstract functional analytic framework that applies to many problems in the existing literature ...

In order to enjoy a digital version of the Jordan Curve Theorem, it is common to use the closed topology for the foreground and the open topology for the background of a 2-dimensional binary image. In this paper, we introduce a single topology that ...

The present paper is devoted to the development of methods and approaches intended for the analysis of natural time series. Due to the strong variability, irregularity, and complex structure of the time series in question, the problem of automatic ...

In this paper we study the shape and growth of structured pseudospectra for small matrix perturbations of the form $A \leadsto A_\Delta=A+B\Delta C$, $\Delta \in \boldsymbol{\Delta}$, $\|\Delta\|\leq \delta$. It is shown that the properly scaled ...

We present a program analysis for verifying quantitative robustness properties of programs, stated generally as: "If the inputs of a program are perturbed by an arbitrary amount epsilon, then its outputs change at most by (K . epsilon), where K can ...

We consider the equation [ r ( t ) x ] + f ( t ) x = 0 as a perturbation of the equation [ r ( t ) y ] + g ( t ) y = 0, where the latter is assumed to be nonoscillatory at infinity. The functions r and g are real-valued, r is ...