Results 81 to 90 of about 85,147 (260)

Improve Adversarial Robustness of AI Models in Remote Sensing via Data-Augmentation and Explainable-AI Methods

open access: yesRemote Sensing
Artificial intelligence (AI) has made remarkable progress in recent years in remote sensing applications, including environmental monitoring, crisis management, city planning, and agriculture.
Sumaiya Tasneem, Kazi Aminul Islam
doaj   +1 more source

Robustness Against Adversarial Attacks Via Learning Confined Adversarial Polytopes

open access: yesICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Comment: The paper has been accepted in ICASSP ...
Hamidi, Shayan Mohajer, Ye, Linfeng
openaire   +2 more sources

Robust Adversarial Reinforcement Learning

open access: yes, 2017
Deep neural networks coupled with fast simulation and improved computation have led to recent successes in the field of reinforcement learning (RL). However, most current RL-based approaches fail to generalize since: (a) the gap between simulation and real world is so large that policy-learning approaches fail to transfer; (b) even if policy learning ...
Pinto, Lerrel   +3 more
openaire   +2 more sources

Low Power Optoelectronic Neuromorphic Memristor for In‐Sensor Computing and Multilevel Hardware Security Communications

open access: yesAdvanced Science, EarlyView.
ABSTRACT Conventional software‐based encryption faces mounting limitations in power efficiency and security, inspiring the development of emerging neuromorphic computing hardware encryption. This study presents a hardware‐level multi‐dimensional encryption paradigm utilizing optoelectronic neuromorphic devices with low energy consumption of 3.3 fJ ...
Bo Sun   +3 more
wiley   +1 more source

Enhancing Adversarial Robustness through Stable Adversarial Training

open access: yesSymmetry
Deep neural network models are vulnerable to attacks from adversarial methods, such as gradient attacks. Evening small perturbations can cause significant differences in their predictions. Adversarial training (AT) aims to improve the model’s adversarial robustness against gradient attacks by generating adversarial samples and optimizing the ...
Kun Yan   +3 more
openaire   +1 more source

Bioinspired Morphology‐Decoupled Soft Gripper with Enhanced Bidirectional Grasping Capability

open access: yesAdvanced Science, EarlyView.
Inspired by biological predation, a novel gripper decouples cross‐modal grasping via dual morphological configurations. Synergistically integrating hybrid rigid and soft coupled fingers with a metamaterial palm, the system performs active compliant grasping for static objects and passive cage capture for dynamic targets.
Yedong Huang   +4 more
wiley   +1 more source

Concealable and Field‐Free Physical Unclonable Function Based on Voltage‐Controlled Magnetic Tunnel Junctions

open access: yesAdvanced Electronic Materials, EarlyView.
A concealable physical unclonable function (PUF) based on an array of 384 nanoscale voltage‐controlled magnetic tunnel junctions is demonstrated. The PUF operates without any external magnetic field. It uses a combination of deterministic and stochastic switching mechanisms, based on the spin transfer torque and voltage‐controlled magnetic anisotropy ...
Thomas Neuner   +6 more
wiley   +1 more source

On the Adversarial Robustness of Hand-Crafted Features and Their Role in Defending Adversarial Examples

open access: yesIEEE Access
Deep Neural Networks (DNNs) have achieved tremendous success in various computer vision tasks but remain highly vulnerable to adversarial examples. To address this limitation, we investigate the inherent robustness of hand-crafted features and validate ...
Shuohan Xue   +2 more
doaj   +1 more source

Pareto adversarial robustness: balancing spatial robustness and sensitivity-based robustness

open access: yesScience China Information Sciences
Adversarial robustness, which primarily comprises sensitivity-based robustness and spatial robustness, plays an integral part in achieving robust generalization. In this paper, we endeavor to design strategies to achieve universal adversarial robustness.
Sun, Ke, Li, Mingjie, Lin, Zhouchen
openaire   +2 more sources

Synthesizing Robust Adversarial Examples

open access: yes, 2017
Standard methods for generating adversarial examples for neural networks do not consistently fool neural network classifiers in the physical world due to a combination of viewpoint shifts, camera noise, and other natural transformations, limiting their relevance to real-world systems.
Athalye, Anish   +3 more
openaire   +2 more sources

Home - About - Disclaimer - Privacy