Multi-task Challenges for Rain Movie Prediction on the Road to Hi-Res Foundation Models

Aleksandra Gruca, Pilar Rípodas, Xavier Calbet, Llorenç Lliso, Federico Serva, Bertrand Le Saux, D P Kreil

Keywords: video frame prediction; transfer learning; super resolution; foundation models; data fusion.

TL;DR: Weather4acst 2024 aim to improve rain forecasts world-wide on an expansive data set with over a magnitude more hi-res rain radar data, allowing a move towards Foundation Models through multi-modality, multi-scale, multi-task challenges

Abstract:

The competition will advance modern algorithms in AI and machine learning through a highly topical interdisciplinary competition challenge: The prediction of hi-res rain radar movies from multi-band satellite sensors requires data fusion of complementary signal sources, multi-channel video frame prediction, as well as super-resolution techniques. To reward models that extract relevant mechanistic patterns reflecting the underlying complex weather systems our evaluation incorporates spatio-temporal shifts: Specifically, algorithms need to forecast several hours of ground-based hi-res precipitation radar from lo-res satellite spectral images in a unique cross-sensor prediction challenge. Models are evaluated within and across regions on Earth with diverse climate and different distributions of heavy precipitation events. Conversely, robustness over time is achieved by testing predictions on data one year after the training period.

Now, in its third year, Weather4acst 2024 aims to improve rain forecasts world-wide on an expansive data set with over a magnitude more hi-res rain radar data, allowing a move towards Foundation Models through multi-modality, multi-scale, multi-task challenges. Accurate rain predictions are becoming ever more critical for everyone, with climate change increasing the frequency of extreme precipitation events. Notably, the new models and insights will have a particular impact for the many regions on Earth where costly weather radar data are not available. Join us on www.weather4cast.net!

Competition Timeline:

18 June 2024 Competition Dataset Release
27 June 2024: STAGE1: Cumulative rainfall Test Set release & Leaderboard open
5 August 2024: STAGE2: Extreme Precipitation Events Test Set release & Leaderboard open
6 October 2024: Test dataset submission deadline.
9 October 2024: Invitation to submit conference abstracts and code.
16 October 2024: Deadline for abstract and code submissions.
25 October 2024: Acceptance notification.

Website: https://weather4cast.net/neurips2024

Primary Contact Email: [e-mail]

Participant Contact Email: [e-mail]

Workshop Format: Virtual (online)

Preferred Timezone: CET (Central European Time)

Submission Number: 40