From Euclidean to Graph-Structured Data : A Survey of Collaborative Learning
The conventional approach to machine learning, that is, collecting data, training models, and performing inference in a single location, faces fundamental limitations, including scalability and privacy, that restrict its applicability.
Fodor, Viktória +2 more
core +1 more source
Tree-Based Machine Learning Models for Predicting Phosphate Adsorption Performance of Layered Double Hydroxides: Insights Into Mechanisms and Process Optimization. [PDF]
Jiang J, Xiong Y.
europepmc +1 more source
An Adaptive Inhibitory WSe2 Transistor for Retinomorphic In‐Sensor Image Processing
Conventional retinomorphic devices typically require deliberate gate‐bias tuning for each illumination condition. In this paper, we demonstrate an adaptive inhibitory WSe2 transistor that uses photo‐induced regime shift between the subthreshold and accumulation modes to achieve decision‐free, intensity‐adaptive image processing within a single pixel ...
Juhwan Baek +12 more
wiley +1 more source
Rapid thermal annealing reconstructs the BCFO/Nb:STO interface into an atomically thin reconstructed interfacial layer, which reshapes the tunneling barrier and converts a high‐TER ferroelectric tunnel junction into an intrinsically nonlinear, self‐rectifying device.
Hojin Lee +21 more
wiley +1 more source
Machine learning-based prediction of SARS-CoV-2 bioactivity: integrating IC<sub>50</sub> regression and activity classification using multi-task neural networks. [PDF]
Maiyza AI, Osama S, Hassan HA.
europepmc +1 more source
Charting Morphotoxicity With Complementary Embryo Models
Three human embryo models of the peri‐implantation stages are exposed to a library of compounds. The (toxic) effects of the compounds are analyzed through automated image analysis pipelines. By evaluating the differential responses of the embryo models, this study underscores the importance of testing complementary embryo models to achieve robust and ...
Dorian G. Luijkx +9 more
wiley +1 more source
Machine Learning and Deep Learning for Earthquake Monitoring: A Systematic Review of Distributed Acoustic Sensing Applications. [PDF]
Iqbal N +4 more
europepmc +1 more source
Silicone breast implants are presented as a model system for understanding polymer permeation in vivo. Rather than representing material failure, “gel bleed” emerges from solution–diffusion transport‐mediated. By integrating polymer architecture, physicochemical transport, and biointerfacial processes, this review provides a unified framework that ...
D. Bouyer +12 more
wiley +1 more source
Identification of novel PKR inhibitory chemotypes <i>via</i> integrated molecular modelling and machine learning. [PDF]
Shakhatreh G, Taha MO, Daoud S.
europepmc +1 more source
Advancing Human Skin Equivalents: The Crucial Role of Neurovascular Integration
This review discusses the importance of integrating vascular and peripheral nerve systems into human skin equivalents (HSEs) to better recapitulate native skin physiology. Recent advances in vascularized, innervated, and neurovascularized HSEs are highlighted, together with emerging bioengineering strategies, current challenges, and future ...
Hao Wu +4 more
wiley +1 more source

