We are happy to share that we have partnered with ELLIS Machine Learning for Earth and Climate Sciences Program 🌍✨!
Read more on our partner page!
Published:
Miki is the Lead Data Scientist at Rewiring America, the leading U.S. nonprofit dedicated to the electrification of American homes. He develops data-driven tools, research, policy, and narratives to help facilitate households' transition from fossil-fuel-powered to efficient electric systems. His recent work includes leading the development of the Residential Electrification Model, a free API that predicts bill savings, emissions reductions, and energy impacts of electrification upgrades at any U.S. address.
Event date:
Webinar with Georges Le Bellier, Conservatoire National des Arts et Métiers. Deep learning for remote sensing plays a crucial role in turning satellite and aerial imagery into dependable, real-world insights. However, Earth observation models must handle diverse environments, sensors, and conditions—such as clouds, seasonal shifts, and geographic differences—while still producing accurate results. In this talk, we explore two paths that lead to more robust and adaptable algorithms: generative domain adaptation and geospatial foundation models. First, I will introduce FlowEO, a generative approach of Unsupervised Domain Adaptation (UDA) for Earth observation, and show its high performance in UDA scenarios for several downstream tasks, including dense prediction and classification. This flow-matching-based translation method improves pretrained predictive models' accuracies in challenging scenarios such as post-disaster response and high cloud coverage cases with SAR-to-optical translation. FlowEO’s generative domain adaptation method is independent of the downstream task and does not require retraining the predictive model. Then, I will present “PANGAEA: A Global and Inclusive Benchmark for Geospatial Foundation Models“, a standardized evaluation protocol that covers a diverse set of datasets, dense prediction tasks, resolutions, sensor modalities, and temporalities. This benchmark includes comparison between geospatial foundation models but also with supervised baselines, namely U-Net and ViT, and highlights the strengths and weaknesses of GFMs. In addition, PANGAEA evaluates models’ accuracy in cases where labels are limited and questions the impact of multi-temporal data for GFMs.
Event date:
Welcome to the official opening of ESA Phi-Lab Sweden in Stockholm. The opening ceremony will take place on August 26, starting with a presentation of the Phi-Lab’s mission and thematic focus, followed by a symbolic ribbon-cutting ceremony. After a light lunch, the afternoon continues with a conference programme featuring inspiring presentations, forward-looking discussions, and a poster session for sharing contributions (pre-registration required). We welcome researchers, start-ups, industry players and the public sector to join this important step towards a future where Sweden plays an active role in shaping AI-driven space applications.
Event date:
The Norwegian government has announced six new national research centers for artificial intelligence, backed by over NOK 1 billion over five years. While their main missions range from education to robotics, creativity, and decision-making, sustainability, ethics, and environmental responsibility are important themes in several. Many PhD positions on relevant topics will be announced in the coming year. The centers will be presented in a public online event on August 26.
