https://pubmed.ncbi.nlm.nih.gov/38091768
Chronic amoxicillin use in rats causes cemental and alveolar bone defects, with decreased calcium and overall density, potentially disturbing tooth attachment integrity.
https://pubmed.ncbi.nlm.nih.gov/38091767
This abstract describes the discovery of rich dynamic behaviors in a new 5D fractional-order memristive neural network (FOMHNN) and its application in image encryption, resulting in a highly efficient and secure encryption scheme with excellent plaintext sensitivity and hardware implementation on FPGA.
https://pubmed.ncbi.nlm.nih.gov/38091766
This study proposes a distributed control protocol for multi-agent systems to solve a distributed time-varying optimization problem with inequality constraints, demonstrating high-efficiency convergence and finite-time consensus for tracking the time-varying global optimal target.
https://pubmed.ncbi.nlm.nih.gov/38091765
This study proposes a distributed deep reinforcement learning method based on a bi-objective framework to improve the ability of multi-robot formation to generalize target position, resulting in better formation maintenance, flexibility, and stability during movement.
https://pubmed.ncbi.nlm.nih.gov/38091764
This study presents a delay-variation-dependent approach for fault detection in discrete-time Markov jump neural networks with time-varying delay and mismatched modes, using an adaptive event-triggered and asynchronous Hโ filter.
https://pubmed.ncbi.nlm.nih.gov/38091763
This study proposes a decentralized learning approach for modeling multi-agent behaviors using partial observation and mechanical constraints, demonstrating improved constraint violations, long-term trajectory prediction, and partial observation in basketball and soccer datasets.
https://pubmed.ncbi.nlm.nih.gov/38091762
The Dominating Set Model Aggregation (DSMA) algorithm, which applies graph theory’s Minimum Connected Dominating Set concept, reduces communication time in decentralized deep learning by up to 100X without sacrificing accuracy, and sometimes even increasing it.
https://pubmed.ncbi.nlm.nih.gov/38091761
This study used massive time-series feature extraction to find that meditation experience is associated with a stability signature in brain dynamics, characterized by higher temporal stability and a specific distributional shape of EEG time-series values.
https://pubmed.ncbi.nlm.nih.gov/38091760
The abstract describes a new point-supervised dense nuclei detection framework that improves weakly-supervised learning for detecting crowded and varying appearance nuclei in histopathological images.
https://pubmed.ncbi.nlm.nih.gov/38091759
This study proposes a neurodynamic penalty approach in the framework of particle swarm optimization to solve a nonconvex distributed optimization problem with nonconvex objective functions and inequality constraints, aiming to find the globally optimal solution through local searches and iterative improvement.