Interested in addressing geographic biases, class imbalances, and presence-only data in species distribution modeling? Don't miss our poster at the Climate Change AI workshop at ICLR 2025. Collaborative work with Nina van Tiel Marc Rußwurm devistuia

🚨🌟 And the ICLR 2025 #ML4RS Best Paper Award 🌟💫 went to Chenhui Zhang (Chenhui Zhang) and Sherrie Wang (Sherrie Wang) for their paper "Good at captioning, bad at counting: Benchmarking GPT-4V on Earth observation data"

🚨🌟 And the ICLR 2025 #ML4RS Best Poster Award 🌟💫 went to Ellianna Abrahams (Ellie Abrahams) with Tasha Snow, Matthew Siegfried, and Fernando Perez (@[email protected]) for their poster "A Concise Tiling Strategy for Preserving Spatial Context in Earth Observation Imagery."

Thrilled that Chenhui Zhang and my work benchmarking vision-language models on Earth observation data was selected for the #ICLR2024 #ML4RS best paper award! Excited to see how foundation models for EO will continue to progress. Check out the paper: arxiv.org/abs/2401.17600

🤖📢 Great topic! Go apply and work with Diego! 😎

Why are we losing tropical forests? 🌴 Together with devistuia, Marc Rußwurm, Jan Dirk Wegner, and others, we used satellite images to map the drivers of forest loss across tropics Read the full paper: iopscience.iop.org/article/10.108…… (1/3)

First great paper from the PHD of Jan Pišl , a pan tropical study of deforestation subsequent land use using crowdsourced data! Collaboration with University of Zurich and IIASA Linda See Gaston Lenczner Lloyd Hughes

Our paper on pseudo-absences in Species Distribution Modeling with Deep Learning is published! sciencedirect.com/science/articl… With Nina van Tiel Benjamin Kellenberger Lloyd Hughes devistuia, we address geographic biases, class imbalance, and presence-only samples in citizen science datasets 🧵

At University of Copenhagen, we are organizing a summer PhD course on SSL4EO. Registration is now open via: ankitkariryaa.github.io/ssl4eo/ (seats are limited). We are looking forward to hear from: Randall Balestriero, Marc Rußwurm, Konstantin Klemmer, Bruno Sánchez-Andrade Nuño, Jan Dirk Wegner, Xiaoxiang ZHU

very happy to talk about perspectives on Earth and climate #FoundationModels at the PhD course on #SSL4EO organized by Nico Lang et al. Other speakers are: Konstantin Klemmer, Jan Dirk Wegner, Marc Rußwurm, Randall Balestriero, Bruno Sánchez-Andrade Nuño, TheoXiong Register here: ankitkariryaa.github.io/ssl4eo/

Excited to be at the SSL4EO summer school in Copenhagen! Marc Rußwurm and I will spend the morning spreading the location encoder gospel - and connecting them more broadly with emerging research on neural implicit representations, e.g. in NeRF, and SSL of Earth representations.🌏🤖

I pre-recorded my talk for the #SSL4EO summer school in Copenhagen this morning. I'm happy to share it here, too. This talk provided the basis for Konstantin Klemmer's presentation and hands-on tutorial on location encoding and #SatCLIP embeddings. Thanks to Machine Learning Section @UCPH Nico Lang

📺New YouTube & Podcast episode: Meta-learning with Meteor, with Marc Rußwurm Meteor showcases the significant potential of few-shot learning for processing remote sensing imagery and proves it's possible to tackle tasks even when very few training examples are available.

Kicking-off the Time-Series Tutorial with Dainius at IEEE IGARSS 2024 📈🛰️💻 Thank you to the remote tutorial contributors Marc Rußwurm Jan Verbesselt

METEOR: Meta-learning to address diverse Earth observation problems across resolutions

Massive Congratulations, Beni 🎉👏