Results 201 to 210 of about 343,225 (391)

A zoogeographic model for the evolution of diversity and endemism in Madagascar

open access: yesEcography, EarlyView.
The delineation of zoogeographic regions is essential for understanding the evolution of biodiversity. Madagascar, characterized by high levels of endemism and habitat diversity, presents unique challenges and opportunities for such studies. Traditional global zoogeographic classifications, largely based on vertebrates, may overlook finer‐scale ...
Gabriela P. Camacho   +3 more
wiley   +1 more source

UPDP: A Unified Progressive Depth Pruner for CNN and Vision Transformer [PDF]

open access: yesarXiv
Traditional channel-wise pruning methods by reducing network channels struggle to effectively prune efficient CNN models with depth-wise convolutional layers and certain efficient modules, such as popular inverted residual blocks. Prior depth pruning methods by reducing network depths are not suitable for pruning some efficient models due to the ...
arxiv  

The effects of climate on bat morphology across space and time

open access: yesEcography, EarlyView.
According to Bergmann's and Allen's rules, climate change may drive morphological shifts in species, affecting body size and appendage length. These rules predict that species in colder climates tend to be larger and have shorter appendages to improve thermoregulation. Bats are thought to be sensitive to climate and are therefore expected to respond to
Laura Paltrinieri   +54 more
wiley   +1 more source

Evolutionary trajectories of multiple defense traits across phylogenetic and geographic scales in Vitis

open access: yesEcography, EarlyView.
The processes driving defense trait correlations may vary within and between species based on ecological or environmental contexts. However, most studies of plant defense theory fail to address this potential for shifts in trait correlations across scales.
Carolyn D. K. Graham, Marjorie G. Weber
wiley   +1 more source

A General Approach to Dropout in Quantum Neural Networks

open access: yesAdvanced Quantum Technologies, EarlyView., 2023
Randomly dropping artificial neurons and all their connections in the training phase reduces overfitting issues in classical neural networks, thus improving performances on previously unseen data. The authors introduce different dropout strategies applied to quantum neural networks, learning models based on parametrized quantum circuits.
Francesco Scala   +3 more
wiley   +1 more source

Home - About - Disclaimer - Privacy