Results 121 to 130 of about 115,711 (260)

Adaptive Autonomy in Microrobot Motion Control via Deep Reinforcement Learning and Path Planning Synergy

Advanced Intelligent Systems, EarlyView.
This study introduces a data‐driven framework that combines deep reinforcement learning with classical path planning to achieve adaptive microrobot navigation. By training a surrogate neural network to emulate microrobot dynamics, the approach improves learning efficiency, reduces training time, and enables robust real‐time obstacle avoidance in ...
Amar Salehi, Bairong Zhu, Haoyu Liu, Tingting Yu   +3 more
wiley   +1 more source

GReAT: A Graph Regularized Adversarial Training Method

IEEE Access
This paper presents GReAT (Graph Regularized Adversarial Training), a novel regularization method designed to enhance the robust classification performance of deep learning models.
Samet Bayram, Kenneth Barner
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

Blind Remote Sensing Image Deblurring Based on Local Maximum High-Frequency Coefficient Prior and Graph Regularization

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
In satellite remote sensing imaging, factors such as optical axis shift, image plane jitter, movement of the target object, and Earth's rotation can induce image blur.
Zhidan Cai, Ming Fang, Zhe Li, Jinyi Ming, Huimin Wang   +4 more
doaj   +1 more source

Survey and Sequence Strategies for Full‐3D Electrical Resistivity Tomography in Archaeological Sites: A Case Study on a Domus of the Roman Town of Augusta Bagiennorum (NW Italy)

Archaeological Prospection, EarlyView.
ABSTRACT Preliminary geophysical investigations are a cost‐effective and efficient way to screen archaeological sites and locate buried structures. Ground‐penetrating radar (GPR) is one of the most widely used methods for archaeological prospection, but in some sites, it cannot be employed effectively due to the presence of clay or other electrically ...
Andrea Vergnano, Diego Franco, Antonio Merico, Francesca Pace, Sofia Uggè, Cesare Comina   +5 more
wiley   +1 more source

Uncertainty Calibration in Molecular Machine Learning: Comparing Evidential and Ensemble Approaches

Chemistry – A European Journal, EarlyView.
ABSTRACT Machine learning (ML) models are increasingly used in quantum chemistry, but their reliability hinges on uncertainty quantification (UQ). In this study, we compare two prominent UQ paradigms—deep evidential regression (DER) and deep ensembles—on the QM9 and WS22 datasets, with a specific emphasis on the role of post hoc calibration.
Bidhan Chandra Garain, Max Pinheiro Jr., Matheus de Oliveira Bispo, Mario Barbatti   +3 more
wiley   +1 more source

Optimal model‐based design of experiments for parameter precision: Supercritical extraction case

The Canadian Journal of Chemical Engineering, EarlyView.
Abstract This study investigates the process of chamomile oil extraction from flowers. A parameter‐distributed model consisting of a set of partial differential equations is used to describe the governing mass transfer phenomena in a cylindrical packed bed with solid chamomile particles under supercritical conditions using carbon dioxide as a solvent ...
Oliwer Sliczniuk, Pekka Oinas
wiley   +1 more source

Restricted Tweedie stochastic block models

Canadian Journal of Statistics, EarlyView.
Abstract The stochastic block model (SBM) is a widely used framework for community detection in networks, where the network structure is typically represented by an adjacency matrix. However, conventional SBMs are not directly applicable to an adjacency matrix that consists of nonnegative zero‐inflated continuous edge weights.
Jie Jian, Mu Zhu, Peijun Sang
wiley   +1 more source

Regular factors in regular graphs

Discrete Mathematics, 1993
The author shows that a \(k\)-regular \((k-1)\)-edge-connected graph with an even number of vertices has an \(m\)-factor not containing \(k-m\) arbitrarily prescribed edges whenever \(1\leq m\leq k-1\). This was known only for \(m=1\) or \(k-1\). Also the sharpness of this result and some corollaries are given.
openaire   +2 more sources

Hidden Markov graphical models with state‐dependent generalized hyperbolic distributions

Canadian Journal of Statistics, EarlyView.
Abstract In this article, we develop a novel hidden Markov graphical model to investigate time‐varying interconnectedness between different financial markets. To identify conditional correlation structures under varying market conditions and accommodate shape features embedded in financial time series, we rely upon the generalized hyperbolic family of ...
Beatrice Foroni, Luca Merlo, Lea Petrella   +2 more
wiley   +1 more source