Results 121 to 130 of about 243,531 (318)
DLA: Dense-Layer-Analysis for Adversarial Example Detection [PDF]
, 2020 Philip Sperl +4 more
openalex +1 more source
Understanding adversarial robustness against on-manifold adversarial examples
Pattern RecognitionDeep neural networks (DNNs) are shown to be vulnerable to adversarial examples. A well-trained model can be easily attacked by adding small perturbations to the original data. One of the hypotheses of the existence of the adversarial examples is the off-manifold assumption: adversarial examples lie off the data manifold. However, recent research showed
Jiancong Xiao +4 more
openaire +2 more sources
Adversarial Examples Are Not Bugs, They Are Features
CoRR, 2019 Adversarial examples have attracted significant attention in machine learning, but the reasons for their existence and pervasiveness remain unclear. We demonstrate that adversarial examples can be directly attributed to the presence of non-robust features: features derived from patterns in the data distribution that are highly predictive, yet brittle ...
Ilyas, A +5 more
openaire +3 more sources
Advanced Intelligent Discovery, EarlyView.Exosomes are emerging as powerful biomarkers for disease diagnosis and monitoring. This review highlights the integration of surface‐enhanced Raman spectroscopy with artificial intelligence to enhance molecular fingerprinting of exosomes. Machine learning and deep learning techniques improve spectral interpretation, enabling accurate classification of ...
Munevver Akdeniz +2 more
wiley +1 more source
A New Kind of Adversarial Example
CoRR, 2022 Almost all adversarial attacks are formulated to add an imperceptible perturbation to an image in order to fool a model. Here, we consider the opposite which is adversarial examples that can fool a human but not a model. A large enough and perceptible perturbation is added to an image such that a model maintains its original decision, whereas a human ...
openaire +2 more sources
Advanced Intelligent Discovery, EarlyView.A novel convolutional neural network architecture enables rapid, unsupervised analysis of IR spectroscopic data from DRIFTS and IRRAS. By combining synthetic data generation with parallel convolutional layers and advanced regularization, the model accurately resolves spectral features of adsorbed CO, offering real‐time insights into ceria surface ...
Mehrdad Jalali +5 more
wiley +1 more source
Image Adversarial Example Generation Method Based on Adaptive Parameter Adjustable Differential Evolution. [PDF]
Entropy (Basel), 2023 Lin Z, Peng C, Tan W, He X.
europepmc +1 more source
Advanced Intelligent Discovery, EarlyView.CrossMatAgent is a multi‐agent framework that combines large language models and diffusion‐based generative AI to automate metamaterial design. By coordinating task‐specific agents—such as describer, architect, and builder—it transforms user‐provided image prompts into high‐fidelity, printable lattice patterns.
Jie Tian +12 more
wiley +1 more source
Robustness Certificates Against Adversarial Examples for ReLU Networks [PDF]
, 2019 Sahil Singla, Soheil Feizi
openalex +1 more source