Results 141 to 150 of about 172,371 (266)

Elastic Fast Marching Learning from Demonstration

Advanced Intelligent Systems, EarlyView.
This article presents Elastic Fast Marching Learning (EFML), a novel approach for learning from demonstration that combines velocity‐based planning with elastic optimization. EFML enables smooth, precise, and adaptable robot trajectories in both position and orientation spaces.
Adrian Prados, Brendan Hertel, Ramon Barber, Reza Azadeh   +3 more
wiley   +1 more source

Securing Generative Artificial Intelligence with Parallel Magnetic Tunnel Junction True Randomness

Advanced Intelligent Systems, EarlyView.
True random numbers can protect generative artificial intelligence (GAI) models from attacks. A highly parallel, spin‐transfer torque magnetic tunnel junction‐based system is demonstrated that generates high‐quality, energy‐efficient random numbers.
Youwei Bao, Shuhan Yang, Hyunsoo Yang
wiley   +1 more source

A CRDNet‐Based Watermarking Algorithm for Fused Visible–Infrared Images

Advanced Intelligent Systems, EarlyView.
CRDnet includes encoders and decoders based on residual and dense structures, a fusion network robust to 12 visible and infrared image fusion algorithms, and predictors for predicting watermarked infrared images. The encoder and decoder incorporate preprocessing steps, attention mechanisms, and activation functions suitable for infrared images.
Yu Bai, Li Li, Shanqing Zhang, Ting Luo, Chin‐Chen Chang   +4 more
wiley   +1 more source

Synthetic Electrocardiogram Spectrogram Generation Using Generative Adversarial Network‐Based Models: A Comparative Study

Advanced Intelligent Systems, EarlyView.
Cardiovascular diseases are leading death causes; electrocardiogram (ECG) analysis is slow, motivating machine learning and deep learning. This study compares deep convolutional generative adversarial network, conditional GAN, and Wasserstein GAN with gradient penalty (WGAN‐GP) for synthetic ECG spectrograms; Fréchet Inception Distance (FID) and ...
Giovanny Barbosa‐Casanova, Norelli Schettini, Winston Percybrooks, Begoña García‐Zapirain   +3 more
wiley   +1 more source

Robust Dysarthric Speech Recognition with GAN Enhancement and LLM Correction

Advanced Intelligent Systems, EarlyView.
This study tackles dysarthric speech recognition by combining generative adversarial network (GAN)‐generated synthetic data with large language model (LLM)‐based error correction. The approach integrates three key elements: an improved CycleGAN to generate synthetic dysarthric speech for data augmentation, a multimodal automatic speech recognition core
Yibo He, Kah Phooi Seng, Chee Shen Lim, Li Minn Ang   +3 more
wiley   +1 more source

Feature Disentangling and Combination Implemented by Spin–Orbit Torque Magnetic Tunnel Junctions

Advanced Intelligent Systems, EarlyView.
Spin–orbit torque magnetic tunnel junctions (SOT‐MTJs) enable efficient feature disentangling and integration in image data. A proposed algorithm leverages SOT‐MTJs as true random number generators to disentangle and recombine features in real time, with experimental validation on emoji and facial datasets.
Xiaohan Li, Yingqian Xu, Caihua Wan, Ran Zhang, Meike Hindenberg, Raik Hoffmann, Shiqiang Liu, Tianyi Zhang, Shilong Xiong, Dehao Kong, Jihao Xia, Fangshuo Gao, Zhuyang Nie, Thomas Kämpfe, Guoqiang Yu, Xiufeng Han   +15 more
wiley   +1 more source

DefenseFea: An Input Transformation Feature Searching Algorithm Based Latent Space for Adversarial Defense

Foundations of Computing and Decision Sciences
Deep neural networks based image classification systems could suffer from adversarial attack algorithms, which generate input examples by adding deliberately crafted yet imperceptible noise to original input images.
Pan Zhang, Yangjie Cao, Chenxi Zhu, Yan Zhuang, Haobo Wang, Jie Li   +5 more
doaj   +1 more source

Upsampling DINOv2 Features for Unsupervised Vision Tasks and Weakly Supervised Materials Segmentation

Advanced Intelligent Systems, EarlyView.
Feature from recent image foundation models (DINOv2) are useful for vision tasks (segmentation, object localization) with little or no human input. Once upsampled, they can be used for weakly supervised micrograph segmentation, achieving strong results when compared to classical features (blurs, edge detection) across a range of material systems.
Ronan Docherty, Antonis Vamvakeros, Samuel J. Cooper   +2 more
wiley   +1 more source

“Will you be there for me?” Social support from family and friends during cold case sexual assault prosecutions

American Journal of Community Psychology, EarlyView.
Abstract If sexual assault survivors report the assault to the criminal legal system, they often need informal support from family and friends throughout the long and frequently retraumatizing process of investigation and prosecution. This study is part of a long‐term community‐based participatory action research project in a predominately Black ...
Rebecca Campbell, Rachael Goodman‐Williams, Katie Gregory, McKenzie Javorka, Jasmine Engleton   +4 more
wiley   +1 more source

Understanding Youth Assaults of Police Officers in Australia: A Power Threat Meaning Framework Analysis

Australian Journal of Social Issues, EarlyView.
ABSTRACT This study explores youth violence towards police officers in Australia through the Power Threat Meaning Framework (PTMF) to better understand the underlying factors contributing to such violence; focusing on power dynamics, childhood adversity, and trauma.
Dimitra Lattas, Kelly Hine, Catherine Creamer, Kelley Burton, Katelyn Davenport‐Klunder   +4 more
wiley   +1 more source