🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Our review "#MachineLearning and #DeepLearning - A review for Ecologists" led by Maximilian Pichler now available as an arXiv.org preprint at arxiv.org/abs/2204.05023. Our paper provides a comprehensive introduction to ML and DL including #Rstats examples.

Our former colleague Theresa Lehmair, now at LfU, is organising a foto competition "Metamorphose Moor". Unfortunately, you have to live in Bavaria to participate, but if that is the case for you, check out the details here lfu.bayern.de/natur/bayaz/ar…

RT Florian Hartig: I recommend checking out the new #RStats package "cito" for building and training neural networks based on #Torch via simple R syntax. Work by our MSc student Christian Amesöder with Maximilian Pichler cran.r-project.org/web/packages/c… Theoretical Ecology Lab @ Uni Regensburg

RT Maximilian Pichler: Douglas Yu @[email protected] Florian Hartig Theoretical Ecology Lab @ Uni Regensburg No, some of functionalities needed in sjSDM are not yet available in rtorch. Also, rtorch does not provide an automatic installation of the GPU dependencies like pytorch does (and the GPU installation is even worse tha…

RT Florian Hartig: Our new study "Climate and parameter sensitivity and induced uncertainties in carbon stock projections for European forests (using LPJ-GUESS 4.0)" led by Johannes Oberpriller now out at GeoModelDevelopment gmd.copernicus.org/articles/15/64… Theoretical Ecology Lab @ Uni Regensburg Brett Lider…

Our new preprint "A Comparison of #DeepLearning Architectures for Inferring Parameters of Diversification Models from Extant Phylogenies" led by Ismaël Lajaaiti w/ Sophia Lambert @JakubVoznica and Helene Morlon now available on bioRxiv biorxiv.org/content/10.110…

Our new preprint " Can predictive models be used for causal inference?" with Florian Hartig is now available as an arXiv.org preprint at arxiv.org/abs/2306.10551. Have you ever wondered if #machinelearning and #deeplearning models can separate collinear effects?

I will soon be opening a 3-yr position for a statistical postdoc / scientific programmer to continue the development of the DHARMa #Rstats #CRAN package for #glmm residual diagnostics. For more info on DHARMa, see cran.r-project.org/web/packages/D…

cito v1.0.2 #rstats package for deep neural networks(#DL #DNN) is now available on #CRAN. New features include confidence intervals and p-values for #xAI metrics: cran.r-project.org/web/packages/c… Florian Hartig Christian Ames

Our review "Novel community data in ecology-properties and prospects" now out Trends in Ecology & Evolution - 50 day free access via authors.elsevier.com/a/1i3CTcZ3WyH4T Joint work of an iDiv Biodiversity Research working group with Jonathan Chase Gurutzeta Guillera alexbush Loïc Pellissier Maximilian Pichler Laura Pollock @ZurellLab

Come work with us - I am looking to fill a 3-yr position for a statistical postdoc / scientific programmer to continue the development of the DHARMa #Rstats #CRAN package for #glmm residual diagnostics. Full job advertisement is here uni-regensburg.de/assets/biologi…

Can complex interactions between trees and their natural enemies shape latitudinal tree diversity patterns in forests? 🌳🌍 Our article in nature delves into this question by comprehensively testing the Janzen-Connell hypothesis using tree mortality data nature.com/articles/s4158…

Our study "Latitudinal patterns in stabilizing density dependence of forest communities" now out in @nature nature.com/articles/s4158… . See also tread by first author Lisa Hülsmann below

cito v1.1 #rstats package for deep neural networks (#DL #DNN) (with formula syntax) is now available on #CRAN. New features include likelihoods such as the negative binomial distribution and easy hyperparameter tuning: cran.r-project.org/web/packages/c… 1/4

Great to see our study on fitting a dynamic range model for birds led by Anne Malchow during her PhD at @ZurellLab now available at Ecological Applications. The #Bayesian Model calibration uses the BayesianTools #Rstats package cran.r-project.org/web/packages/B…

Our new study on a #Bayesian calibrating of #treeRegeneration in a process-based forest model using the BayesianTools #rstats package now out at GeoModelDevelopment. Led by Yannek Käber with Harald Bugmann at Forest Ecology ETH. For a link to the study, see below the tweet of the first author!

In our Ecography software note ‘cito': an R package for training neural networks using ‘torch' (#CRAN #rstats) led by Christian Ames and together with Florian Hartig we present a new package for fitting and interpreting #DNN (#DL) doi.org/10.1111/ecog.0… Theoretical Ecology Lab @ Uni Regensburg 1/5

Our article "The road to integrate climate change projections with regional land‐use–biodiversity models" led by Juliano Sarmento Cabral with many others including now published in People and Nature dx.doi.org/10.1002/pan3.1…

Great to see our article "Consequences of Local Conspecific Density Effects for Plant Diversity and Community Dynamics" lead by Joe LaManna and Camille Delavaux now out and available in Ecology Letters onlinelibrary.wiley.com/doi/ftr/10.111…

Our preprint "Calibration, sensitivity and uncertainty analysis of ecological models -- a review" with Anne Malchow now available at doi.org/10.22541/au.17… This is our attempt at giving a broad overview about methods and issues surrounding