The diversity bonus in pooling local knowledge about complex problems
… were more closely matched scientific expert opinions. These … Random experiments also
predicted that knowledge diversity … We first combine individual maps by stakeholder types to …
predicted that knowledge diversity … We first combine individual maps by stakeholder types to …
[HTML][HTML] Global patterns and drivers of bee distribution
… checklist of bee species distributions and >5,800,000 public … turn, use these predictions to
model bee richness worldwide. … For listing and a map of the countries that fall into each region…
model bee richness worldwide. … For listing and a map of the countries that fall into each region…
[HTML][HTML] Crop mapping from image time series: Deep learning with multi-scale label hierarchies
… expert knowledge and significantly improves the mapping of … Formally, our objective is to
predict a crop type map Y ∈ I R H×… To match that format, we assign the same label to all pixels …
predict a crop type map Y ∈ I R H×… To match that format, we assign the same label to all pixels …
Predicting conversion to wet age-related macular degeneration using deep learning
… automatic tissue maps, our system predicts conversion to exAMD … expert performance on
the task, we compare model … referral and diagnosis model closely matched the performance …
the task, we compare model … referral and diagnosis model closely matched the performance …
[HTML][HTML] Topography and human pressure in mountain ranges alter expected species responses to climate change
… limit species distributions and movement required to shift their … were made by expert evaluation
using maps, atlases, and … This resulted in a seamless elevation dataset that matched the …
using maps, atlases, and … This resulted in a seamless elevation dataset that matched the …
Macroevolutionary convergence connects morphological form to ecological function in birds
… The position of species within this space maps with 70–85% accuracy onto major niche
axes… phenotypically matched pairs of families by fitting a random forest model predicting family …
axes… phenotypically matched pairs of families by fitting a random forest model predicting family …
Stochastic segmentation networks: Modelling spatially correlated aleatoric uncertainty
M Monteiro, L Le Folgoc… - Advances in neural …, 2020 - proceedings.neurips.cc
… , and it is not possible to match an annotation to an expert. Thus, the four sets of … model,
we used the mean of the logit map distribution. We computed the DSC between the prediction …
we used the mean of the logit map distribution. We computed the DSC between the prediction …
Birds, bats and beyond: Evaluating generalization in bioacoustics models
… enthusiasts with a wide range of expertise. While many recordings … a one-to-one mapping
between predictions and ground truth … When an event predicted by the model is matched to a …
between predictions and ground truth … When an event predicted by the model is matched to a …
[HTML][HTML] Tree species classification from airborne hyperspectral and LiDAR data using 3D convolutional neural networks
… After segmenting the full study area, we matched the delineated … The tree species distributions
of our training and validation … We aim to use the tree species maps produced in this study …
of our training and validation … We aim to use the tree species maps produced in this study …
[HTML][HTML] Restoring and attributing ancient texts using deep neural networks
… work shows how matching human experts with deep learning … distributions for all tasks using
multiple charts and maps, and … measure the distance in years between a prediction and the …
multiple charts and maps, and … measure the distance in years between a prediction and the …